Name
stringlengths 3
106
| Subsets
list | HF Link
stringlengths 3
69
| Link
stringlengths 24
135
| License
stringclasses 28
values | Year
int32 2k
2.02k
| Language
stringclasses 2
values | Dialect
stringclasses 18
values | Domain
stringclasses 10
values | Form
stringclasses 3
values | Collection Style
stringclasses 7
values | Description
stringlengths 16
1.64k
| Volume
stringlengths 1
13
| Unit
stringclasses 5
values | Ethical Risks
stringclasses 4
values | Provider
stringlengths 1
136
| Derived From
stringlengths 2
307
| Paper Title
stringlengths 3
143
| Paper Link
stringlengths 3
285
| Script
stringclasses 4
values | Tokenized
stringclasses 2
values | Host
stringclasses 21
values | Access
stringclasses 3
values | Cost
stringlengths 3
11
| Test Split
stringclasses 3
values | Tasks
stringlengths 8
181
| Venue Title
stringlengths 2
46
| Citations
stringlengths 3
6
| Venue Type
stringclasses 5
values | Venue Name
stringlengths 3
113
| Authors
stringlengths 3
923
| Affiliations
stringlengths 1
470
| Abstract
stringlengths 3
2.15k
| Added By
stringlengths 3
25
|
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
Shami | [
{
"Name": "Jordanian",
"Dialect": "ar-JO: (Arabic (Jordan))",
"Volume": "32,078",
"Unit": "sentences"
},
{
"Name": "Palestanian",
"Dialect": "ar-PS: (Arabic (Palestine))",
"Volume": "21,264",
"Unit": "sentences"
},
{
"Name": "Syrian",
"Dialect": "ar-SY: (Arabic (Syria))",
"Volume": "48,159",
"Unit": "sentences"
},
{
"Name": "Lebanese",
"Dialect": "ar-LB: (Arabic (Lebanon))",
"Volume": "16,304",
"Unit": "sentences"
}
] | https://huggingface.co/datasets/arbml/Shami | https://github.com/GU-CLASP/shami-corpus | Apache-2.0 | 2,018 | ar | ar-LEV: (Arabic(Levant)) | social media | text | crawling and annotation(other) | the first Levantine Dialect Corpus (SDC) covering data from the four dialects spoken in Palestine, Jordan, Lebanon and Syria. | 117,805 | sentences | Medium | Multiple institutions | nan | Shami: A Corpus of Levantine Arabic Dialects | https://aclanthology.org/L18-1576.pdf | Arab | No | GitHub | Free | nan | No | dialect identification | LREC | 25.0 | conference | International Conference on Language Resources and Evaluation | Chatrine Qwaider,Motaz Saad,S. Chatzikyriakidis,Simon Dobnik | ,The Islamic University of Gaza,, | Modern Standard Arabic (MSA) is the official language used in education and media across the Arab world both in writing and formal speech. However, in daily communication several dialects depending on the country, region as well as other social factors, are used. With the emergence of social media, the dialectal amount of data on the Internet have increased and the NLP tools that support MSA are not well-suited to process this data due to the difference between the dialects and MSA. In this paper, we construct the Shami corpus, the first Levantine Dialect Corpus (SDC) covering data from the four dialects spoken in Palestine, Jordan, Lebanon and Syria. We also describe rules for pre-processing without affecting the meaning so that it is processable by NLP tools. We choose Dialect Identification as the task to evaluate SDC and compare it with two other corpora. In this respect, experiments are conducted using different parameters based on n-gram models and Naive Bayes classifiers. SDC is larger than the existing corpora in terms of size, words and vocabularies. In addition, we use the performance on the Language Identification task to exemplify the similarities and differences in the individual dialects. | nan |
LABR | [] | https://huggingface.co/datasets/labr | https://github.com/mohamedadaly/LABR | GPL-2.0 | 2,013 | ar | mixed | reviews | text | crawling and annotation(other) | The largest sentiment analysis dataset to-date for the Arabic language. | 63,257 | sentences | Low | Cairo University | nan | LABR: A Large Scale Arabic Book Reviews Dataset | https://aclanthology.org/P13-2088.pdf | Arab | No | GitHub | Free | nan | Yes | sentiment analysis | ACL | 165.0 | conference | Associations of computation linguistics | Mohamed A. Aly,A. Atiya | , | We introduce LABR, the largest sentiment analysis dataset to-date for the Arabic language. It consists of over 63,000 book reviews, each rated on a scale of 1 to 5 stars. We investigate the properties of the the dataset, and present its statistics. We explore using the dataset for two tasks: sentiment polarity classification and rating classification. We provide standard splits of the dataset into training and testing, for both polarity and rating classification, in both balanced and unbalanced settings. We run baseline experiments on the dataset to establish a benchmark. | nan |
Arabic POS Dialect | [
{
"Name": "Egyptian",
"Dialect": "ar-EG: (Arabic (Egypt))",
"Volume": "350",
"Unit": "sentences"
},
{
"Name": "Levantine",
"Dialect": "ar-LEV: (Arabic(Levant))",
"Volume": "350",
"Unit": "sentences"
},
{
"Name": "Gulf",
"Dialect": "ar-GLF: (Arabic (Gulf))",
"Volume": "350",
"Unit": "sentences"
},
{
"Name": "Maghrebi",
"Dialect": "ar-MA: (Arabic (Morocco))",
"Volume": "350",
"Unit": "sentences"
}
] | https://huggingface.co/datasets/arabic_pos_dialect | https://github.com/qcri/dialectal_arabic_resources | unknown | 2,018 | ar | mixed | social media | text | crawling and annotation(other) | includes tweets in Egyptian, Levantine, Gulf, and Maghrebi, with 350 tweets for each dialect with appropriate train/test/development splits for 5-fold cross validation | 1,400 | sentences | Medium | QCRI | nan | Multi-Dialect Arabic POS Tagging: A CRF Approach
| https://aclanthology.org/L18-1015.pdf | Arab | Yes | GitHub | Free | nan | No | part of speech tagging | LREC | 17.0 | conference | International Conference on Language Resources and Evaluation | Kareem Darwish,Hamdy Mubarak,Ahmed Abdelali,M. Eldesouki,Younes Samih,Randah Alharbi,Mohammed Attia,Walid Magdy,Laura Kallmeyer | ,,,,University Of Düsseldorf;Computational Linguistics,,,The University of Edinburgh, | This paper introduces a new dataset of POS-tagged Arabic tweets in four major dialects along with tagging guidelines. The data, which we are releasing publicly, includes tweets in Egyptian, Levantine, Gulf, and Maghrebi, with 350 tweets for each dialect with appropriate train/test/development splits for 5-fold cross validation. We use a Conditional Random Fields (CRF) sequence labeler to train POS taggers for each dialect and examine the effect of cross and joint dialect training, and give benchmark results for the datasets. Using clitic n-grams, clitic metatypes, and stem templates as features, we were able to train a joint model that can correctly tag four different dialects with an average accuracy of 89.3%. | nan |
Emotional-Tone | [] | https://huggingface.co/datasets/emotone_ar | https://github.com/AmrMehasseb/Emotional-Tone | unknown | 2,017 | ar | ar-EG: (Arabic (Egypt)) | social media | text | crawling and annotation(other) | emotion detection dataset | 10,065 | sentences | Medium | Nile University | nan | Emotional Tone Detection in Arabic Tweets | https://www.researchgate.net/profile/Samhaa-El-Beltagy/publication/320271778_Emotional_Tone_Detection_in_Arabic_Tweets/links/59d9f0a5458515a5bc2b1d8a/Emotional-Tone-Detection-in-Arabic-Tweets.pdf | Arab | No | GitHub | Free | nan | No | emotion classification | CICLing | 10.0 | conference | International Conference on Computational Linguistics and Intelligent Text Processing | Amr Al-Khatib,S. El-Beltagy | , | Emotion detection in Arabic text is an emerging research area, but the efforts in this new field have been hindered by the very limited availability of Arabic datasets annotated with emotions. In this paper, we review work that has been carried out in the area of emotion analysis in Arabic text. We then present an Arabic tweet dataset that we have built to serve this task. The efforts and methodologies followed to collect, clean, and annotate our dataset are described and preliminary experiments carried out on this dataset for emotion detection are presented. The results of these experiments are provided as a benchmark for future studies and comparisons with other emotion detection models. The best results over a set of eight emotions were obtained using a complement Naive Bayes algorithm with an overall accuracy of 68.12%. | nan |
OCLAR | [] | https://huggingface.co/datasets/oclar | http://archive.ics.uci.edu/ml/datasets/Opinion+Corpus+for+Lebanese+Arabic+Reviews+%28OCLAR%29# | unknown | 2,019 | ar | ar-LB: (Arabic (Lebanon)) | reviews | text | crawling and annotation(other) | Opinion Corpus for Lebanese Arabic Reviews | 3,916 | sentences | Low | Lebanese University | nan | Sentiment Classifier: Logistic Regression for Arabic Services’ Reviews in Lebanon | https://ieeexplore.ieee.org/abstract/document/8716394/ | Arab | No | other | Free | nan | No | sentiment analysis, review classification | ICCIS | 8.0 | conference | International Conference on Computer and Information Sciences | Marwan Al Omari,Moustafa Al-Hajj,N. Hammami,A. Sabra | Université de Poitiers,,, | This paper proposes a logistic regression approach paired with term and inverse document frequency (TF*IDF) for Arabic sentiment classification on services’ reviews in Lebanon country. Reviews are about public services, including hotels, restaurants, shops, and others. We collected manually from Google reviews and Zomato, which have reached to 3916 reviews. Experiments show three core findings: 1) The classifier is confident when used to predict positive reviews. 2) The model is biased on predicting reviews with negative sentiment. Finally, the low percentage of negative reviews in the corpus contributes to the diffidence of logistic regression model. | nan |
Commonsense validation | [] | https://huggingface.co/datasets/arbml/Commonsense_Validation | https://github.com/msmadi/Arabic-Dataset-for-Commonsense-Validationion | CC BY-SA 4.0 | 2,020 | ar | ar-MSA: (Arabic (Modern Standard Arabic)) | other | text | machine translation | a benchmark Arabic dataset for commonsense understanding and validation | 12,000 | sentences | Low | Jordan University | nan | Is this sentence valid? An Arabic Dataset for Commonsense Validation
| https://arxiv.org/abs/2008.10873 | Arab | No | GitHub | Free | nan | Yes | commonsense validation | ArXiv | 1.0 | preprint | ArXiv | Saja Khaled Tawalbeh,Mohammad Al-Smadi | , | The commonsense understanding and validation remains a challenging task in the field of natural language understanding. Therefore, several research papers have been published that studied the capability of proposed systems to evaluate the models ability to validate commonsense in text. In this paper, we present a benchmark Arabic dataset for commonsense understanding and validation as well as a baseline research and models trained using the same dataset. To the best of our knowledge, this dataset is considered as the first in the field of Arabic text commonsense validation. The dataset is distributed under the Creative Commons BY-SA 4.0 license and can be found on GitHub. | nan |
SANAD | [] | https://huggingface.co/datasets/arbml/SANAD | https://data.mendeley.com/datasets/57zpx667y9/2 | CC BY 4.0 | 2,019 | ar | ar-MSA: (Arabic (Modern Standard Arabic)) | news articles | text | crawling and annotation(other) | textual data collected from three news portals | 194,797 | documents | Low | Sharjah University | nan | SANAD: Single-label Arabic News Articles Dataset for automatic text categorization
| https://www.sciencedirect.com/science/article/pii/S2352340919304305 | Arab | No | Mendeley Data | Free | nan | Yes | topic classification | Data in brief | 18.0 | journal | Data in brief | Omar Einea,Ashraf Elnagar,Ridhwan Al Debsi | ,, | Text Classification is one of the most popular Natural Language Processing (NLP) tasks. Text classification (aka categorization) is an active research topic in recent years. However, much less attention was directed towards this task in Arabic, due to the lack of rich representative resources for training an Arabic text classifier. Therefore, we introduce a large Single-labeled Arabic News Articles Dataset (SANAD) of textual data collected from three news portals. The dataset is a large one consisting of almost 200k articles distributed into seven categories that we offer to the research community on Arabic computational linguistics. We anticipate that this rich dataset would make a great aid for a variety of NLP tasks on Modern Standard Arabic (MSA) textual data, especially for single label text classification purposes. We present the data in raw form. SANAD is composed of three main datasets scraped from three news portals, which are AlKhaleej, AlArabiya, and Akhbarona. SANAD is made public and freely available at https://data.mendeley.com/datasets/57zpx667y9. | nan |
BRAD 1.0 | [] | https://huggingface.co/datasets/arbml/BRAD | https://github.com/elnagara/BRAD-Arabic-Dataset | unknown | 2,016 | ar | mixed | reviews | text | crawling and annotation(other) | The reviews were collected from GoodReads.com website during June/July 2016 | 156,506 | sentences | Low | Sharjah University | nan | BRAD 1.0: Book reviews in Arabic dataset
| https://ieeexplore.ieee.org/abstract/document/7945800 | Arab | No | GitHub | Free | nan | Yes | review classification | AICCSA | 32.0 | conference | International Conference on Computer Systems and Applications | Ashraf Elnagar,Omar Einea | , | The availability of rich datasets is a pre-requisite for proposing robust sentiment analysis systems. A variety of such datasets exists in English language. However, it is rare or nonexistent for the Arabic language except for a recent LABR dataset, which consists of a little bit over 63,000 book reviews extracted from. Goodreads. com. We introduce BRAD 1.0, the largest Book Reviews in Arabic Dataset for sentiment analysis and machine language applications. BRAD comprises of almost 510,600 book records. Each record corresponds for a single review and has the review in Arabic language and the reviewer's rating on a scale of 1 to 5 stars. In this paper, we present and describe the properties of BRAD. Further, we provide two versions of BRAD: the complete unbalanced dataset and the balanced version of BRAD. Finally, we implement four sentiment analysis classifiers based on this dataset and report our findings. When training and testing the classifiers on BRAD as opposed to LABR, an improvement rate growth of 46% is reported. The highest accuracy attained is 91%. Our core contribution is to make this benchmark-dataset available and accessible to the research community on Arabic language. | nan |
ArCOV-19 | [] | https://huggingface.co/datasets/ar_cov19 | https://gitlab.com/bigirqu/ArCOV-19 | unknown | 2,021 | ar | mixed | social media | text | crawling and annotation(other) | Arabic COVID-19 Twitter dataset that covers the period from 27th of January till 5th of May 2021. | 3,140,158 | sentences | Medium | Qatar University | nan | ArCOV-19: The First Arabic COVID-19 Twitter Dataset
with Propagation Networks | https://camel.abudhabi.nyu.edu/WANLP-2021-Program/47_Paper.pdf | Arab | Yes | GitLab | Free | nan | No | information retrieval,social computing | WANLP | 18.0 | workshop | Arabic Natural Language Processing Workshop | Fatima Haouari,Maram Hasanain,Reem Suwaileh,T. Elsayed | ,,, | In this paper, we present ArCOV-19, an Arabic COVID-19 Twitter dataset that spans one year, covering the period from 27th of January 2020 till 31st of January 2021. ArCOV-19 is the first publicly-available Arabic Twitter dataset covering COVID-19 pandemic that includes about 2.7M tweets alongside the propagation networks of the most-popular subset of them (i.e., most-retweeted and -liked). The propagation networks include both retweetsand conversational threads (i.e., threads of replies). ArCOV-19 is designed to enable research under several domains including natural language processing, information retrieval, and social computing. Preliminary analysis shows that ArCOV-19 captures rising discussions associated with the first reported cases of the disease as they appeared in the Arab world.In addition to the source tweets and the propagation networks, we also release the search queries and the language-independent crawler used to collect the tweets to encourage the curation of similar datasets. | nan |
Gumar | [
{
"Name": "SA",
"Dialect": "ar-SA: (Arabic (Saudi Arabia))",
"Volume": "748",
"Unit": "documents"
},
{
"Name": "AE",
"Dialect": "ar-AE: (Arabic (United Arab Emirates))",
"Volume": "165",
"Unit": "documents"
},
{
"Name": "KW",
"Dialect": "ar-KW: (Arabic (Kuwait))",
"Volume": "73",
"Unit": "documents"
},
{
"Name": "OM",
"Dialect": "ar-OM: (Arabic (Oman))",
"Volume": "14",
"Unit": "documents"
},
{
"Name": "QA",
"Dialect": "ar-QA: (Arabic (Qatar))",
"Volume": "8",
"Unit": "documents"
},
{
"Name": "BH",
"Dialect": "ar-BH: (Arabic (Bahrain))",
"Volume": "6",
"Unit": "documents"
},
{
"Name": "GA",
"Dialect": "ar-GLF: (Arabic (Gulf))",
"Volume": "123",
"Unit": "documents"
},
{
"Name": "Arabic",
"Dialect": "ar-MSA: (Arabic (Modern Standard Arabic))",
"Volume": "98",
"Unit": "documents"
}
] | nan | https://camel.abudhabi.nyu.edu/gumar/?page=download&lang=en | custom | 2,016 | ar | mixed | other | text | crawling and annotation(other) | a large-scale corpus of Gulf Arabic consisting of 110 million words from 1,200 forum novels | 1,236 | documents | Low | NYU Abu Dhabi | nan | A Large Scale Corpus of Gulf Arabic
| https://aclanthology.org/L16-1679.pdf | Arab | No | CAMeL Resources | Upon-Request | nan | No | morphological analysis | LREC | 37.0 | conference | International Conference on Language Resources and Evaluation | Salam Khalifa,Nizar Habash,D. Abdulrahim,Sara Hassan | New York University Abu Dhabi,,, | Most Arabic natural language processing tools and resources are developed to serve Modern Standard Arabic (MSA), which is the official written language in the Arab World. Some Dialectal Arabic varieties, notably Egyptian Arabic, have received some attention lately and have a growing collection of resources that include annotated corpora and morphological analyzers and taggers. Gulf Arabic, however, lags behind in that respect. In this paper, we present the Gumar Corpus, a large-scale corpus of Gulf Arabic consisting of 110 million words from 1,200 forum novels. We annotate the corpus for sub-dialect information at the document level. We also present results of a preliminary study in the morphological annotation of Gulf Arabic which includes developing guidelines for a conventional orthography. The text of the corpus is publicly browsable through a web interface we developed for it. | nan |
Arab-Acquis | [] | nan | https://camel.abudhabi.nyu.edu/arabacquis/ | custom | 2,017 | multilingual | ar-MSA: (Arabic (Modern Standard Arabic)) | other | text | crawling and annotation(translation) | consists of over 12,000 sentences from the JRCAcquis (Acquis Communautaire) corpus | 12,000 | sentences | Low | NYU Abu Dhabi | nan | A Parallel Corpus for Evaluating Machine Translation between Arabic and European Languages | https://aclanthology.org/E17-2038.pdf | Arab | No | CAMeL Resources | Upon-Request | nan | Yes | machine translation | EACL | 9.0 | conference | European Chapter of the Association for Computational Linguistics | Nizar Habash,Nasser Zalmout,Dima Taji,Hieu Hoang,Maverick Alzate | ,,,, | We present Arab-Acquis, a large publicly available dataset for evaluating machine translation between 22 European languages and Arabic. Arab-Acquis consists of over 12,000 sentences from the JRC-Acquis (Acquis Communautaire) corpus translated twice by professional translators, once from English and once from French, and totaling over 600,000 words. The corpus follows previous data splits in the literature for tuning, development, and testing. We describe the corpus and how it was created. We also present the first benchmarking results on translating to and from Arabic for 22 European languages. | nan |
MADAR | [] | nan | https://camel.abudhabi.nyu.edu/madar-parallel-corpus/ | custom | 2,018 | ar | mixed | other | text | manual curation | a collection of parallel sentences covering the dialects of 25 cities from the Arab World | 14,000 | sentences | Low | NYU Abu Dhabi | nan | The MADAR Arabic Dialect Corpus and Lexicon
| http://www.lrec-conf.org/proceedings/lrec2018/pdf/351.pdf | Arab | No | CAMeL Resources | Upon-Request | nan | No | dialect identification | LREC | 85.0 | conference | International Conference on Language Resources and Evaluation | Houda Bouamor,Nizar Habash,Mohammad Salameh,W. Zaghouani,Owen Rambow,D. Abdulrahim,Ossama Obeid,Salam Khalifa,Fadhl Eryani,Alexander Erdmann,Kemal Oflazer | ,,,,,,,New York University Abu Dhabi,,, | In this paper, we present two resources that were created as part of the Multi Arabic Dialect Applications and Resources (MADAR) project. The first is a large parallel corpus of 25 Arabic city dialects in the travel domain. The second is a lexicon of 1,045 concepts with an average of 45 words from 25 cities per concept. These resources are the first of their kind in terms of the breadth of their coverage and the fine location granularity. The focus on cities, as opposed to regions in studying Arabic dialects, opens new avenues to many areas of research from dialectology to dialect identification and machine translation. | nan |
HARD | [] | https://huggingface.co/datasets/hard | https://github.com/elnagara/HARD-Arabic-Dataset | unknown | 2,018 | ar | mixed | reviews | text | crawling | 490587 hotel reviews collected from the Booking.com website. | 93,700 | sentences | Low | Sharjah University | nan | Hotel Arabic-Reviews Dataset Construction for Sentiment Analysis Applications | https://link.springer.com/chapter/10.1007/978-3-319-67056-0_3 | Arab | No | GitHub | Free | nan | No | sentiment analysis, review classification | INLP | 49.0 | journal | Intelligent Natural Language Processing: Trends and Applications | Ashraf Elnagar,Yasmin Khalifa,Anas Einea | ,, | Arabic language suffers from the lack of available large datasets for machine learning and sentiment analysis applications. This work adds to the recently reported large dataset BRAD, which is the largest Book Reviews in Arabic Dataset. In this paper, we introduce HARD (Hotel Arabic-Reviews Dataset), the largest Book Reviews in Arabic Dataset for subjective sentiment analysis and machine language applications. HARD comprises of 490587 hotel reviews collected from the Booking.com website. Each record contains the review text in the Arabic language, the reviewer’s rating on a scale of 1 to 10 stars, and other attributes about the hotel/reviewer. We make available the full unbalanced dataset as well as a balanced subset. To examine the datasets, we implement six popular classifiers using Modern Standard Arabic (MSA) as well as Dialectal Arabic (DA). We test the sentiment analyzers for polarity and rating classifications. Furthermore, we implement a polarity lexicon-based sentiment analyzer. The findings confirm the effectiveness of the classifiers and the datasets. Our core contribution is to make this benchmark-dataset available and accessible to the research community on Arabic language. | nan |
Let-mi | [] | nan | https://github.com/bilalghanem/let-mi | unknown | 2,021 | ar | ar-LEV: (Arabic(Levant)) | social media | text | crawling and annotation(other) | Levantine Twitter dataset for Misogynistic language | 6,603 | sentences | Low | Multiple institutions | nan | Let-Mi: An Arabic Levantine Twitter Dataset for Misogynistic Language
| https://arxiv.org/pdf/2103.10195.pdf | Arab | No | other | Upon-Request | nan | No | misogyny identification | WANLP | 2.0 | workshop | Arabic Natural Language Processing Workshop | Hala Mulki,Bilal Ghanem | , | Online misogyny has become an increasing worry for Arab women who experience gender-based online abuse on a daily basis. Misogyny automatic detection systems can assist in the prohibition of anti-women Arabic toxic content. Developing such systems is hindered by the lack of the Arabic misogyny benchmark datasets. In this paper, we introduce an Arabic Levantine Twitter dataset for Misogynistic language (LeT-Mi) to be the first benchmark dataset for Arabic misogyny. We further provide a detailed review of the dataset creation and annotation phases. The consistency of the annotations for the proposed dataset was emphasized through inter-rater agreement evaluation measures. Moreover, Let-Mi was used as an evaluation dataset through binary/multi-/target classification tasks conducted by several state-of-the-art machine learning systems along with Multi-Task Learning (MTL) configuration. The obtained results indicated that the performances achieved by the used systems are consistent with state-of-the-art results for languages other than Arabic, while employing MTL improved the performance of the misogyny/target classification tasks. | nan |
Aljazeera-dialectal speech | [] | nan | https://alt.qcri.org/resources/aljazeeraspeechcorpus/ | unknown | 2,015 | ar | mixed | transcribed audio | spoken | other | utterance-level dialect labels for 57 hours of high-quality audio from Al Jazeera consisting of four major varieties of DA: Egyptian, Levantine, Gulf, and North African. | 57 | hours | Low | QCRI | nan | Crowdsource a little to label a lot:
Labeling a Speech Corpus of Dialectal Arabic | https://www.isca-speech.org/archive/interspeech_2015/papers/i15_2824.pdf | Arab | No | QCRI Resources | Free | nan | No | speech recognition | INTERSPEECH | 23.0 | conference | Conference of the International Speech Communication Association | Samantha Wray,Ahmed Ali | , | Arabic is a language with great dialectal variety, with Modern Standard Arabic (MSA) being the only standardized dialect. Spoken Arabic is characterized by frequent code-switching between MSA and Dialectal Arabic (DA). DA varieties are typically differentiated by region, but despite their wide-spread usage, they are under-resourced and lack viable corpora and tools necessary for speech recognition and natural language processing. Existing DA speech corpora are limited in scope, consisting of mainly telephone conversations and scripted speech. In this paper we describe our efforts for using crowdsourcing to create a labeled multi-dialectal speech corpus. We obtained utterance-level dialect labels for 57 hours of high-quality audio from Al Jazeera consisting of four major varieties of DA: Egyptian, Levantine, Gulf, and North African. Using speaker linking to identify utterances spoken by the same speaker, and measures of label accuracy likelihood based on annotator behavior, we automatically labeled an additional 94 hours. The complete corpus contains 850 hours with approximately 18% DA speech. | nan |
AJGT | [] | https://huggingface.co/datasets/ajgt_twitter_ar | https://github.com/komari6/Arabic-twitter-corpus-AJGT | unknown | 2,017 | ar | ar-JO: (Arabic (Jordan)) | social media | text | crawling and annotation(other) | Corpus consisted of 1,800 tweets annotated as positive and negative. Modern Standard Arabic (MSA) or Jordanian dialect. | 1,800 | sentences | Medium | Multiple institutions | nan | Arabic Tweets Sentimental Analysis Using Machine Learning | https://link.springer.com/chapter/10.1007/978-3-319-60042-0_66 | Arab | No | GitHub | Free | nan | No | sentiment analysis | IEA/AIE | 53.0 | conference | International Conference on Industrial, Engineering and Other Applications of Applied Intelligent Systems | K. Alomari,H. M. Elsherif,K. Shaalan | ,, | The continuous rapid growth of electronic Arabic contents in social media channels and in Twitter particularly poses an opportunity for opinion mining research. Nevertheless, it is hindered by either the lack of sentimental analysis resources or Arabic language text analysis challenges. This study introduces an Arabic Jordanian twitter corpus where Tweets are annotated as either positive or negative. It investigates different supervised machine learning sentiment analysis approaches when applied to Arabic user’s social media of general subjects that are found in either Modern Standard Arabic (MSA) or Jordanian dialect. Experiments are conducted to evaluate the use of different weight schemes, stemming and N-grams terms techniques and scenarios. The experimental results provide the best scenario for each classifier and indicate that SVM classifier using term frequency–inverse document frequency (TF-IDF) weighting scheme with stemming through Bigrams feature outperforms the Naive Bayesian classifier best scenario performance results. Furthermore, this study results outperformed other results from comparable related work. | nan |
Arabic Speech Corpus | [] | https://huggingface.co/datasets/arabic_speech_corpus | http://en.arabicspeechcorpus.com/ | CC BY 4.0 | 2,015 | ar | ar-MSA: (Arabic (Modern Standard Arabic)) | transcribed audio | spoken | manual curation | The corpus was recorded in south Levantine Arabic (Damascian accent) using a professional studio. | 4 | hours | Low | SOUTHAMPTON University | nan | Modern Standard Arabic Speech Corpus
| https://ota.bodleian.ox.ac.uk/repository/xmlui/bitstream/handle/20.500.12024/2561/arabic-speech-corpus-report.pdf?sequence=3 | Arab-Latn | No | other | Free | nan | No | speech recognition | other | 3.0 | preprint | nan | Nawar Halabi,Gary B Wills | , | Corpus design for speech synthesis is a well-researched topic in languages such as English
compared to Modern Standard Arabic, and there is a tendency to focus on methods to automatically
generate the orthographic transcript to be recorded (usually greedy methods), which was used in
this work. In this work, a study of Modern Standard Arabic (MSA) phonetics and phonology is
conducted in order to develop criteria for a greedy method to create a MSA speech corpus
transcript for recording. The size of the dataset is reduced a number of times using optimisation
methods with different parameters to yield a much smaller dataset with the identical phonetic
coverage offered before the reduction. The resulting output transcript is then chosen for recording.
A phoneme set and a phonotactic rule-set are created for automatically generating a phonetic
transcript of normalised MSA text which is used to annotate and segment the speech corpus after
recording, achieving 82.5% boundary precision with some manual alignments (~15% of the
corpus) to increase the precision of the automatic alignment. This is part of a larger work to create
a completely annotated and segmented speech corpus for MSA speech synthesis with an evaluation
of the quality of this speech corpus and, where possible, the quality of each stage in the process.
| nan |
ArSarcasm | [
{
"Name": "Egyptian",
"Dialect": "ar-EG: (Arabic (Egypt))",
"Volume": "2,383",
"Unit": "sentences"
},
{
"Name": "Gulf",
"Dialect": "ar-GLF: (Arabic (Gulf))",
"Volume": "519",
"Unit": "sentences"
},
{
"Name": "Levantine",
"Dialect": "ar-LEV: (Arabic(Levant))",
"Volume": "551",
"Unit": "sentences"
},
{
"Name": "Maghrebi",
"Dialect": "ar-MA: (Arabic (Morocco))",
"Volume": "32",
"Unit": "sentences"
},
{
"Name": "MSA",
"Dialect": "ar-MSA: (Arabic (Modern Standard Arabic))",
"Volume": "7,062",
"Unit": "sentences"
}
] | https://huggingface.co/datasets/ar_sarcasm | https://github.com/iabufarha/ArSarcasm | unknown | 2,020 | ar | mixed | social media | text | crawling and annotation(other) | The dataset was created using previously available Arabic sentiment analysis datasets | 8,437 | sentences | Low | Multiple institutions | ASTD | From Arabic Sentiment Analysis to Sarcasm Detection:
The ArSarcasm Dataset | https://aclanthology.org/2020.osact-1.5.pdf | Arab | No | GitHub | Free | nan | Yes | dialect identification, sentiment analysis, sarcasm detection | OSACT | 23.0 | workshop | Workshop on Open-Source Arabic Corpora and Processing Tools | Ibrahim Abu Farha,Walid Magdy | University of Edinburgh,The University of Edinburgh | Sarcasm is one of the main challenges for sentiment analysis systems. Its complexity comes from the expression of opinion using implicit indirect phrasing. In this paper, we present ArSarcasm, an Arabic sarcasm detection dataset, which was created through the reannotation of available Arabic sentiment analysis datasets. The dataset contains 10,547 tweets, 16% of which are sarcastic. In addition to sarcasm the data was annotated for sentiment and dialects. Our analysis shows the highly subjective nature of these tasks, which is demonstrated by the shift in sentiment labels based on annotators’ biases. Experiments show the degradation of state-of-the-art sentiment analysers when faced with sarcastic content. Finally, we train a deep learning model for sarcasm detection using BiLSTM. The model achieves an F1 score of 0.46, which shows the challenging nature of the task, and should act as a basic baseline for future research on our dataset. | nan |
ArSentiment | [] | https://huggingface.co/datasets/ar_res_reviews | https://github.com/hadyelsahar/large-arabic-sentiment-analysis-resouces | unknown | 2,015 | ar | mixed | reviews | text | crawling | Automatically annotated Reviews in Domains of Movies, Hotels, Restaurants and Products | 42,692 | sentences | Low | Nile University | nan | Building Large Arabic Multi-domain Resources for Sentiment Analysis | https://link.springer.com/chapter/10.1007/978-3-319-18117-2_2 | Arab | No | GitHub | Free | nan | No | sentiment analysis, review classification | CICLing | 127.0 | conference | International Conference on Computational Linguistics and Intelligent Text Processing | Hady ElSahar,S. El-Beltagy | , | While there has been a recent progress in the area of Arabic Sentiment Analysis, most of the resources in this area are either of limited size, domain specific or not publicly available. In this paper, we address this problem by generating large multi-domain datasets for Sentiment Analysis in Arabic. The datasets were scrapped from different reviewing websites and consist of a total of 33K annotated reviews for movies, hotels, restaurants and products. Moreover we build multi-domain lexicons from the generated datasets. Different experiments have been carried out to validate the usefulness of the datasets and the generated lexicons for the task of sentiment classification. From the experimental results, we highlight some useful insights addressing: the best performing classifiers and feature representation methods, the effect of introducing lexicon based features and factors affecting the accuracy of sentiment classification in general. All the datasets, experiments code and results have been made publicly available for scientific purposes. | nan |
AMARA | [] | https://huggingface.co/datasets/qed_amara | https://alt.qcri.org/resources/qedcorpus/ | custom | 2,014 | multilingual | ar-MSA: (Arabic (Modern Standard Arabic)) | other | text | crawling and annotation(translation) | multilingually aligned for 20 languages, i.e. 20 monolingual corpora and 190 parallel corpora | 154,301 | sentences | Low | QCRI | nan | The AMARA Corpus: Building Parallel Language Resources for the
Educational Domain | http://www.lrec-conf.org/proceedings/lrec2014/pdf/877_Paper.pdf | Arab | No | QCRI Resources | Free | nan | Yes | machine translation | LREC | 59.0 | conference | International Conference on Language Resources and Evaluation | Ahmed Abdelali,Francisco Guzmán,Hassan Sajjad,S. Vogel | ,,, | This paper presents the AMARA corpus of on-line educational content: a new parallel corpus of educational video subtitles, multilingually aligned for 20 languages, i.e. 20 monolingual corpora and 190 parallel corpora. This corpus includes both resource-rich languages such as English and Arabic, and resource-poor languages such as Hindi and Thai. In this paper, we describe the gathering, validation, and preprocessing of a large collection of parallel, community-generated subtitles. Furthermore, we describe the methodology used to prepare the data for Machine Translation tasks. Additionally, we provide a document-level, jointly aligned development and test sets for 14 language pairs, designed for tuning and testing Machine Translation systems. We provide baseline results for these tasks, and highlight some of the challenges we face when building machine translation systems for educational content. | nan |
MKQA | [] | https://huggingface.co/datasets/mkqa | https://github.com/apple/ml-mkqa | CC BY-SA 3.0 | 2,020 | multilingual | mixed | other | text | human translation | 10k question-answer pairs aligned across 26 typologically diverse languages | 10,000 | sentences | Low | Apple | Natural Questions | MKQA: A Linguistically Diverse Benchmark for
Multilingual Open Domain Question Answering
| https://arxiv.org/pdf/2007.15207.pdf | Arab | No | GitHub | Free | nan | Yes | question answering | ArXiv | 11.0 | preprint | ArXiv | S. Longpre,Yi Lu,Joachim Daiber | ,,Apple | Progress in cross-lingual modeling depends on challenging, realistic, and diverse evaluation sets. We introduce Multilingual Knowledge Questions and Answers (MKQA), an open-domain question answering evaluation set comprising 10k question-answer pairs aligned across 26 typologically diverse languages (260k question-answer pairs in total). The goal of this dataset is to provide a challenging benchmark for question answering quality across a wide set of languages. Answers are based on a language-independent data representation, making results comparable across languages and independent of language-specific passages. With 26 languages, this dataset supplies the widest range of languages to-date for evaluating question answering. We benchmark state-of-the-art extractive question answering baselines, trained on Natural Questions, including Multilingual BERT, and XLM-RoBERTa, in zero shot and translation settings. Results indicate this dataset is challenging, especially in low-resource languages. | nan |
journalists_questions | [] | https://huggingface.co/datasets/journalists_questions | http://qufaculty.qu.edu.qa/telsayed/datasets/ | unknown | 2,016 | ar | mixed | social media | text | human translation | crowdsorucing to collect
binary annotations for 10K of the potential question tweets
based on whether they truly contain questions or not | 10,000 | sentences | Medium | Qatar University | nan | What Questions Do Journalists Ask on Twitter? | https://www.semanticscholar.org/paper/What-Questions-Do-Journalists-Ask-on-Twitter-Hasanain-Bagdouri/d1b32df7e9f39e6fba912cc209054ae0256638eb | Arab | No | Dropbox | Free | nan | No | question answering | ICWSM | 4.0 | conference | International Conference on Web and Social Media | Maram Hasanain,Mossaab Bagdouri,T. Elsayed,D. Oard | ,,, | Social media platforms are a major source of information for both the general public and for journalists. Journalists use Twitter and other social media services to gather story ideas, to find eyewitnesses, and for a wide range of other purposes. One way in which journalists use Twitter is to ask questions. This paper reports on an empirical investigation of questions asked by Arab journalists on Twitter. The analysis begins with the development of an ontology of question types, proceeds to human annotation of training and test data, and concludes by reporting the level of accuracy that can be achieved with automated classification techniques. The results show good classifier effectiveness for high prevalence question types, but that obtaining sufficient training data for lower prevalence question types can be challenging. | nan |
arabic billion words | [] | https://huggingface.co/datasets/arabic_billion_words | http://www.abuelkhair.net/index.php/en/arabic/abu-el-khair-corpus | unknown | 2,016 | ar | ar-MSA: (Arabic (Modern Standard Arabic)) | news articles | text | crawling | includes more than five million newspaper articles | 5,222,973 | documents | Low | Multiple institutions | nan | 1.5 billion words Arabic Corpus
| https://arxiv.org/ftp/arxiv/papers/1611/1611.04033.pdf | Arab | No | other | Free | nan | No | text generation, language modeling | ArXiv | 17.0 | preprint | ArXiv | I. A. El-Khair | nan | This study is an attempt to build a contemporary linguistic corpus for Arabic language. The corpus produced, is a text corpus includes more than five million newspaper articles. It contains over a billion and a half words in total, out of which, there is about three million unique words. The data were collected from newspaper articles in ten major news sources from eight Arabic countries, over a period of fourteen years. The corpus was encoded with two types of encoding, namely: UTF-8, and Windows CP-1256. Also it was marked with two mark-up languages, namely: SGML, and XML. | nan |
AraCOVID19-MFH | [] | nan | https://github.com/MohamedHadjAmeur/AraCOVID19-MFH | CC BY-NC-SA 4.0 | 2,021 | ar | mixed | social media | text | crawling and annotation(other) | multi-label fake news and hate speech detection dataset each sentence is annotated with 10 labels | 10,828 | sentences | High | Multiple institutions | nan | AraCOVID19-MFH: Arabic COVID-19 Multi-label Fake News and Hate Speech Detection Dataset | https://arxiv.org/abs/2105.03143 | Arab | No | GitHub | Upon-Request | nan | No | fake news detection, hate speech detection | ArXiv | 0.0 | preprint | ArXiv | Mohamed Seghir Hadj Ameur,H. Aliane | , | Along with the COVID-19 pandemic, an "infodemic" of false and misleading information has emerged and has complicated the COVID-19 response efforts. Social networking sites such as Facebook and Twitter have contributed largely to the spread of rumors, conspiracy theories, hate, xenophobia, racism, and prejudice. To combat the spread of fake news, researchers around the world have and are still making considerable efforts to build and share COVID-19 related research articles, models, and datasets. This paper releases "AraCOVID19-MFH"1a manually annotated multi-label Arabic COVID-19 fake news and hate speech detection dataset. Our dataset contains 10,828 Arabic tweets annotated with 10 different labels. The labels have been designed to consider some aspects relevant to the fact-checking task, such as the tweet's check worthiness, positivity/negativity, and factuality. To confirm our annotated dataset's practical utility, we used it to train and evaluate several classification models and reported the obtained results. Though the dataset is mainly designed for fake news detection, it can also be used for hate speech detection, opinion/news classification, dialect identification, and many other tasks. © 2021 Elsevier B.V.. All rights reserved. | nan |
Arabic Dialects | [
{
"Name": "Algerian",
"Dialect": "ar-DZ: (Arabic (Algeria))",
"Volume": "13,393",
"Unit": "sentences"
},
{
"Name": "Lebanon",
"Dialect": "ar-LB: (Arabic (Lebanon))",
"Volume": "14,482",
"Unit": "sentences"
},
{
"Name": "Morocco",
"Dialect": "ar-MA: (Arabic (Morocco))",
"Volume": "9,965",
"Unit": "sentences"
},
{
"Name": "Tunisian",
"Dialect": "ar-TN: (Arabic (Tunisia))",
"Volume": "8,044",
"Unit": "sentences"
},
{
"Name": "Egyptian",
"Dialect": "ar-EG: (Arabic (Egypt))",
"Volume": "7,519",
"Unit": "sentences"
}
] | nan | https://msda.um6p.ma/msda_datasets | unknown | 2,021 | ar | mixed | social media | text | crawling and annotation(other) | 50K tweets labeled for multiple tasks | 50,000 | sentences | Medium | Mohamed VI Polytechnic University | nan | An open access NLP dataset for Arabic dialects :
data collection, labeling, and model construction | https://arxiv.org/pdf/2102.11000.pdf | Arab | No | other | Free | nan | No | dialect identification, topic classification, sentiment analysis | ArXiv | 0.0 | preprint | ArXiv | ElMehdi Boujou,Hamza Chataoui,Abdellah El Mekki,Saad Benjelloun,I. Chairi,Ismail Berrada | ,,Mohammed VI Polytechnic University,,, | Natural Language Processing (NLP) is today a very active field of research and innovation. Many applications need however big sets of data for supervised learning, suitably labelled for the training purpose. This includes applications for the Arabic language and its national dialects. However, such open access labeled data sets in Arabic and its dialects are lacking in the Data Science ecosystem and this lack can be a burden to innovation and research in this field. In this work, we present an open data set of social data content in several Arabic dialects. This data was collected from the Twitter social network and consists on +50K twits in five (5) national dialects. Furthermore, this data was labeled for several applications, namely dialect detection, topic detection and sentiment analysis. We publish this data as an open access data to encourage innovation and encourage other works in the field of NLP for Arabic dialects and social media. A selection of models were built using this data set and are presented in this paper along with their performances. | nan |
OSIAN | [] | nan | http://oujda-nlp-team.net/en/corpora/osian-corpus/ | CC BY-NC 4.0 | 2,019 | ar | mixed | news articles | text | crawling | The corpus data was collected from international Arabic news websites, | 3,500,000 | documents | Low | Multiple institutions | nan | OSIAN: Open Source International Arabic News Corpus -
Preparation and Integration into the CLARIN-infrastructure
| https://aclanthology.org/W19-4619.pdf | Arab | No | other | Free | nan | No | text generation, language modeling | WANLP | 15.0 | workshop | Arabic Natural Language Processing Workshop | Imad Zeroual,Dirk Goldhahn,Thomas Eckart,A. Lakhouaja | ,,, | The World Wide Web has become a fundamental resource for building large text corpora. Broadcasting platforms such as news websites are rich sources of data regarding diverse topics and form a valuable foundation for research. The Arabic language is extensively utilized on the Web. Still, Arabic is relatively an under-resourced language in terms of availability of freely annotated corpora. This paper presents the first version of the Open Source International Arabic News (OSIAN) corpus. The corpus data was collected from international Arabic news websites, all being freely available on the Web. The corpus consists of about 3.5 million articles comprising more than 37 million sentences and roughly 1 billion tokens. It is encoded in XML; each article is annotated with metadata information. Moreover, each word is annotated with lemma and part-of-speech. the described corpus is processed, archived and published into the CLARIN infrastructure. This publication includes descriptive metadata via OAI-PMH, direct access to the plain text material (available under Creative Commons Attribution-Non-Commercial 4.0 International License - CC BY-NC 4.0), and integration into the WebLicht annotation platform and CLARIN’s Federated Content Search FCS. | nan |
ArabicWeb16 | [] | nan | https://sites.google.com/view/arabicweb16/ | CC BY 3.0 | 2,016 | ar | mixed | other | text | crawling | public Web crawl of 150,211,934 Arabic Web pages with high coverage of dialectal Arabic as well as Modern Standard Arabic (MSA) | 150,211,934 | documents | Low | Qatar University | nan | ArabicWeb16: A New Crawl for Today’s Arabic Web
| https://www.ischool.utexas.edu/~ml/papers/sigir16-arabicweb.pdf | Arab | No | google sites | Upon-Request | nan | No | text generation, language modeling | SIGIR | 12.0 | conference | ACM SIGIR Conference on Research and Development in Information Retrieval | Reem Suwaileh,Mucahid Kutlu,Nihal Fathima,T. Elsayed,Matthew Lease | ,TOBB University of Economics and Technology,,, | Web crawls provide valuable snapshots of the Web which enable a wide variety of research, be it distributional analysis to characterize Web properties or use of language, content analysis in social science, or Information Retrieval (IR) research to develop and evaluate effective search algorithms. While many English-centric Web crawls exist, existing public Arabic Web crawls are quite limited, limiting research and development. To remedy this, we present ArabicWeb16, a new public Web crawl of roughly 150M Arabic Web pages with significant coverage of dialectal Arabic as well as Modern Standard Arabic. For IR researchers, we expect ArabicWeb16 to support various research areas: ad-hoc search, question answering, filtering, cross-dialect search, dialect detection, entity search, blog search, and spam detection. Combined use with a separate Arabic Twitter dataset we are also collecting may provide further value. | nan |
Arabic OSCAR | [] | https://huggingface.co/datasets/oscar | https://oscar-corpus.com/ | CC0 | 2,020 | ar | mixed | other | text | crawling | a huge multilingual corpus obtained by language classification and filtering of the Common Crawl | 8,117,162,828 | tokens | Low | Inria | Common Crawl | A Monolingual Approach to Contextualized Word Embeddings
for Mid-Resource Languages | https://arxiv.org/pdf/2006.06202.pdf | Arab | No | other | Free | nan | No | text generation, language modeling | ACL | 39.0 | conference | Assofications of computation linguisitcs | Pedro Javier Ortiz Suárez,L. Romary,Benoît Sagot | Inria;Sorbonne Université,, | We use the multilingual OSCAR corpus, extracted from Common Crawl via language classification, filtering and cleaning, to train monolingual contextualized word embeddings (ELMo) for five mid-resource languages. We then compare the performance of OSCAR-based and Wikipedia-based ELMo embeddings for these languages on the part-of-speech tagging and parsing tasks. We show that, despite the noise in the Common-Crawl-based OSCAR data, embeddings trained on OSCAR perform much better than monolingual embeddings trained on Wikipedia. They actually equal or improve the current state of the art in tagging and parsing for all five languages. In particular, they also improve over multilingual Wikipedia-based contextual embeddings (multilingual BERT), which almost always constitutes the previous state of the art, thereby showing that the benefit of a larger, more diverse corpus surpasses the cross-lingual benefit of multilingual embedding architectures. | nan |
Tashkeela | [
{
"Name": "Classical Arabic",
"Dialect": "ar-CLS: (Arabic (Classic))",
"Volume": "74,762,008",
"Unit": "tokens"
},
{
"Name": "Modern Standard Arabic",
"Dialect": "ar-MSA: (Arabic (Modern Standard Arabic))",
"Volume": "867,913",
"Unit": "tokens"
},
{
"Name": "Manual",
"Dialect": "mixed",
"Volume": "7,701",
"Unit": "tokens"
}
] | https://huggingface.co/datasets/tashkeela | https://sourceforge.net/projects/tashkeela/ | GPL-2.0 | 2,017 | ar | mixed | books | text | crawling | Arabic discritization Corpus, Resource, Arabic vocalized texts | 75,629,921 | tokens | Low | ESI | nan | Tashkeela: Novel corpus of Arabic vocalized texts, data for auto-diacritization systems | https://www.sciencedirect.com/science/article/pii/S2352340917300112 | Arab | No | sourceforge | Free | nan | No | diacritization | Data in brief | 46.0 | journal | Data in brief | Taha Zerrouki,Amar Balla | , | Arabic diacritics are often missed in Arabic scripts. This feature is a handicap for new learner to read َArabic, text to speech conversion systems, reading and semantic analysis of Arabic texts. The automatic diacritization systems are the best solution to handle this issue. But such automation needs resources as diactritized texts to train and evaluate such systems. In this paper, we describe our corpus of Arabic diacritized texts. This corpus is called Tashkeela. It can be used as a linguistic resource tool for natural language processing such as automatic diacritics systems, dis-ambiguity mechanism, features and data extraction. The corpus is freely available, it contains 75 million of fully vocalized words mainly 97 books from classical and modern Arabic language. The corpus is collected from manually vocalized texts using web crawling process. | nan |
MGB-2 | [] | nan | https://arabicspeech.org/mgb2/ | unknown | 2,017 | ar | ar-MSA: (Arabic (Modern Standard Arabic)) | transcribed audio | spoken | crawling and annotation(other) | from Aljazeera TV programs have been manually captioned with no timing information | 1,200 | hours | Low | QCRI | nan | SPEECH RECOGNITION CHALLENGE IN THE WILD: ARABIC MGB-3
| https://arxiv.org/pdf/1709.07276.pdf | Arab | Yes | other | Upon-Request | nan | No | speech recognition | ASRU | 64.0 | workshop | IEEE Automatic Speech Recognition and Understanding Workshop | A. Ali,S. Vogel,S. Renals | ,, | This paper describes the Arabic MGB-3 Challenge — Arabic Speech Recognition in the Wild. Unlike last year's Arabic MGB-2 Challenge, for which the recognition task was based on more than 1,200 hours broadcast TV news recordings from Aljazeera Arabic TV programs, MGB-3 emphasises dialectal Arabic using a multi-genre collection of Egyptian YouTube videos. Seven genres were used for the data collection: comedy, cooking, family/kids, fashion, drama, sports, and science (TEDx). A total of 16 hours of videos, split evenly across the different genres, were divided into adaptation, development and evaluation data sets. The Arabic MGB-Challenge comprised two tasks: A) Speech transcription, evaluated on the MGB-3 test set, along with the 10 hour MGB-2 test set to report progress on the MGB-2 evaluation; B) Arabic dialect identification, introduced this year in order to distinguish between four major Arabic dialects — Egyptian, Levantine, North African, Gulf, as well as Modern Standard Arabic. Two hours of audio per dialect were released for development and a further two hours were used for evaluation. For dialect identification, both lexical features and i-vector bottleneck features were shared with participants in addition to the raw audio recordings. Overall, thirteen teams submitted ten systems to the challenge. We outline the approaches adopted in each system, and summarise the evaluation results. | nan |
MGB-3 | [] | nan | https://arabicspeech.org/mgb3-asr-2/ | unknown | 2,017 | ar | ar-EG: (Arabic (Egypt)) | transcribed audio | spoken | crawling and annotation(other) | explores multi-genre data; comedy, cooking, cultural, environment, family-kids, fashion, movies-drama, sports, and science talks (TEDX) | 16 | hours | Low | QCRI | nan | SPEECH RECOGNITION CHALLENGE IN THE WILD: ARABIC MGB-3
| https://arxiv.org/pdf/1709.07276.pdf | Arab | Yes | other | Upon-Request | nan | No | speech recognition | ASRU | 64.0 | workshop | IEEE Automatic Speech Recognition and Understanding Workshop | A. Ali,S. Vogel,S. Renals | ,, | This paper describes the Arabic MGB-3 Challenge — Arabic Speech Recognition in the Wild. Unlike last year's Arabic MGB-2 Challenge, for which the recognition task was based on more than 1,200 hours broadcast TV news recordings from Aljazeera Arabic TV programs, MGB-3 emphasises dialectal Arabic using a multi-genre collection of Egyptian YouTube videos. Seven genres were used for the data collection: comedy, cooking, family/kids, fashion, drama, sports, and science (TEDx). A total of 16 hours of videos, split evenly across the different genres, were divided into adaptation, development and evaluation data sets. The Arabic MGB-Challenge comprised two tasks: A) Speech transcription, evaluated on the MGB-3 test set, along with the 10 hour MGB-2 test set to report progress on the MGB-2 evaluation; B) Arabic dialect identification, introduced this year in order to distinguish between four major Arabic dialects — Egyptian, Levantine, North African, Gulf, as well as Modern Standard Arabic. Two hours of audio per dialect were released for development and a further two hours were used for evaluation. For dialect identification, both lexical features and i-vector bottleneck features were shared with participants in addition to the raw audio recordings. Overall, thirteen teams submitted ten systems to the challenge. We outline the approaches adopted in each system, and summarise the evaluation results. | nan |
MGB-5 | [] | nan | https://arabicspeech.org/mgb5/ | unknown | 2,019 | ar | ar-MA: (Arabic (Morocco)) | transcribed audio | spoken | crawling and annotation(other) | Moroccan Arabic speech extracted from 93 YouTube videos distributed across seven genres: comedy, cooking, family/children, fashion, drama, sports, and science clips. | 14 | hours | Low | QCRI | nan | The MGB-5 Challenge: Recognition and Dialect Identification of Dialectal Arabic Speech | https://ieeexplore.ieee.org/document/9003960 | Arab | Yes | other | Upon-Request | nan | No | speech recognition | ASRU | 18.0 | workshop | IEEE Automatic Speech Recognition and Understanding Workshop | A. Ali,Suwon Shon,Younes Samih,Hamdy Mubarak,Ahmed Abdelali,James R. Glass,S. Renals,K. Choukri | ,,University Of Düsseldorf;Computational Linguistics,,,,, | This paper describes the fifth edition of the Multi-Genre Broadcast Challenge (MGB-5), an evaluation focused on Arabic speech recognition and dialect identification. MGB-5 extends the previous MGB-3 challenge in two ways: first it focuses on Moroccan Arabic speech recognition; second the granularity of the Arabic dialect identification task is increased from 5 dialect classes to 17, by collecting data from 17 Arabic speaking countries. Both tasks use YouTube recordings to provide a multi-genre multi-dialectal challenge in the wild. Moroccan speech transcription used about 13 hours of transcribed speech data, split across training, development, and test sets, covering 7-genres: comedy, cooking, family/kids, fashion, drama, sports, and science (TEDx). The fine-grained Arabic dialect identification data was collected from known YouTube channels from 17 Arabic countries. 3,000 hours of this data was released for training, and 57 hours for development and testing. The dialect identification data was divided into three sub-categories based on the segment duration: short (under 5 s), medium (5–20 s), and long (>20 s). Overall, 25 teams registered for the challenge, and 9 teams submitted systems for the two tasks. We outline the approaches adopted in each system and summarize the evaluation results. | nan |
ADI-5 | [
{
"Name": "Egyptian",
"Dialect": "ar-EG: (Arabic (Egypt))",
"Volume": "14.4",
"Unit": "hours"
},
{
"Name": "Gulf",
"Dialect": "ar-GLF: (Arabic (Gulf))",
"Volume": "14.1",
"Unit": "hours"
},
{
"Name": "Levantine",
"Dialect": "ar-LEV: (Arabic(Levant))",
"Volume": "14.3",
"Unit": "hours"
},
{
"Name": "MSA",
"Dialect": "ar-MSA: (Arabic (Modern Standard Arabic))",
"Volume": "14.3",
"Unit": "hours"
},
{
"Name": "North African",
"Dialect": "ar-NOR: (Arabic (North Africa))",
"Volume": "14.6",
"Unit": "hours"
}
] | nan | https://arabicspeech.org/mgb3-adi/ | unknown | 2,016 | ar | mixed | transcribed audio | spoken | crawling and annotation(other) | This will be divided across the five major Arabic dialects; Egyptian (EGY), Levantine (LAV), Gulf (GLF), North African (NOR), and Modern Standard Arabic (MSA) | 50 | hours | Low | QCRI | nan | Automatic Dialect Detection in Arabic Broadcast Speech
| https://arxiv.org/pdf/1509.06928.pdf | Arab | No | other | Upon-Request | nan | No | dialect identification | INTERSPEECH | 93.0 | conference | Conference of the International Speech Communication Association | A. Ali,Najim Dehak,P. Cardinal,Sameer Khurana,S. Yella,James R. Glass,P. Bell,S. Renals | ,,,,,,, | We investigate different approaches for dialect identification in Arabic broadcast speech, using phonetic, lexical features obtained from a speech recognition system, and acoustic features using the i-vector framework. We studied both generative and discriminate classifiers, and we combined these features using a multi-class Support Vector Machine (SVM). We validated our results on an Arabic/English language identification task, with an accuracy of 100%. We used these features in a binary classifier to discriminate between Modern Standard Arabic (MSA) and Dialectal Arabic, with an accuracy of 100%. We further report results using the proposed method to discriminate between the five most widely used dialects of Arabic: namely Egyptian, Gulf, Levantine, North African, and MSA, with an accuracy of 52%. We discuss dialect identification errors in the context of dialect code-switching between Dialectal Arabic and MSA, and compare the error pattern between manually labeled data, and the output from our classifier. We also release the train and test data as standard corpus for dialect identification. | nan |
QASR | [] | nan | https://arabicspeech.org/qasr/ | unknown | 2,021 | ar | mixed | transcribed audio | spoken | crawling and annotation(other) | This multi-dialect speech dataset contains 2, 000 hours of speech sampled at 16kHz crawled from Aljazeera news channel | 2,000 | hours | Low | QCRI | nan | QASR: QCRI Aljazeera Speech Resource
A Large Scale Annotated Arabic Speech Corpus | https://arxiv.org/pdf/2106.13000.pdf | Arab | Yes | other | Upon-Request | nan | No | speech recognition | ACL | 2.0 | conference | Assofications of computation linguisitcs | Hamdy Mubarak,Amir Hussein,S. A. Chowdhury | ,, | We introduce the largest transcribed Arabic speech corpus, QASR1, collected from the broadcast domain. This multi-dialect speech dataset contains 2, 000 hours of speech sampled at 16kHz crawled from Aljazeera news channel. The dataset is released with lightly supervised transcriptions, aligned with the audio segments. Unlike previous datasets, QASR contains linguistically motivated segmentation, punctuation, speaker information among others. QASR is suitable for training and evaluating speech recognition systems, acousticsand/or linguisticsbased Arabic dialect identification, punctuation restoration, speaker identification, speaker linking, and potentially other NLP modules for spoken data. In addition to QASR transcription, we release a dataset of 130M words to aid in designing and training a better language model. We show that end-to-end automatic speech recognition trained on QASR reports a competitive word error rate compared to the previous MGB-2 corpus. We report baseline results for downstream natural language processing tasks such as named entity recognition using speech transcript. We also report the first baseline for Arabic punctuation restoration. We make the corpus available for the research community. | nan |
ADI-17 | [
{
"Name": "Algeria",
"Dialect": "ar-DZ: (Arabic (Algeria))",
"Volume": "115.7",
"Unit": "hours"
},
{
"Name": "Egypt",
"Dialect": "ar-EG: (Arabic (Egypt))",
"Volume": "451.1",
"Unit": "hours"
},
{
"Name": "Iraq",
"Dialect": "ar-IQ: (Arabic (Iraq))",
"Volume": "815.8",
"Unit": "hours"
},
{
"Name": "Jordan",
"Dialect": "ar-JO: (Arabic (Jordan))",
"Volume": "25.9",
"Unit": "hours"
},
{
"Name": "Saudi Arabia",
"Dialect": "ar-SA: (Arabic (Saudi Arabia))",
"Volume": "186.1",
"Unit": "hours"
},
{
"Name": "Kuwait",
"Dialect": "ar-KW: (Arabic (Kuwait))",
"Volume": "108.2",
"Unit": "hours"
},
{
"Name": "Lebanon",
"Dialect": "ar-LB: (Arabic (Lebanon))",
"Volume": "116.8",
"Unit": "hours"
},
{
"Name": "Libya",
"Dialect": "ar-LY: (Arabic (Libya))",
"Volume": "127.4",
"Unit": "hours"
},
{
"Name": "Mauritania",
"Dialect": "ar-MR: (Arabic (Mauritania))",
"Volume": "456.4",
"Unit": "hours"
},
{
"Name": "Morocco",
"Dialect": "ar-MA: (Arabic (Morocco))",
"Volume": "57.8",
"Unit": "hours"
},
{
"Name": "Oman",
"Dialect": "ar-OM: (Arabic (Oman))",
"Volume": "58.5",
"Unit": "hours"
},
{
"Name": "Palestine",
"Dialect": "ar-PS: (Arabic (Palestine))",
"Volume": "121.4",
"Unit": "hours"
},
{
"Name": "Qatar",
"Dialect": "ar-QA: (Arabic (Qatar))",
"Volume": "62.3",
"Unit": "hours"
},
{
"Name": "Sudan",
"Dialect": "ar-SD: (Arabic (Sudan))",
"Volume": "47.7",
"Unit": "hours"
},
{
"Name": "Syria",
"Dialect": "ar-SY: (Arabic (Syria))",
"Volume": "119.5",
"Unit": "hours"
},
{
"Name": "UAE",
"Dialect": "ar-AE: (Arabic (United Arab Emirates))",
"Volume": "108.4",
"Unit": "hours"
},
{
"Name": "Yemen",
"Dialect": "ar-YE: (Arabic (Yemen))",
"Volume": "53.4",
"Unit": "hours"
}
] | nan | https://arabicspeech.org/mgb5/#adi17 | unknown | 2,019 | ar | mixed | transcribed audio | spoken | crawling and annotation(other) | dialect identification of speech from YouTube to one of the 17 dialects | 3,091 | hours | Low | QCRI | nan | The MGB-5 Challenge: Recognition and Dialect Identification of Dialectal Arabic Speech | https://ieeexplore.ieee.org/document/9003960 | Arab | No | other | Upon-Request | nan | Yes | dialect identification | ASRU | 18.0 | workshop | IEEE Automatic Speech Recognition and Understanding Workshop | A. Ali,Suwon Shon,Younes Samih,Hamdy Mubarak,Ahmed Abdelali,James R. Glass,S. Renals,K. Choukri | ,,University Of Düsseldorf;Computational Linguistics,,,,, | This paper describes the fifth edition of the Multi-Genre Broadcast Challenge (MGB-5), an evaluation focused on Arabic speech recognition and dialect identification. MGB-5 extends the previous MGB-3 challenge in two ways: first it focuses on Moroccan Arabic speech recognition; second the granularity of the Arabic dialect identification task is increased from 5 dialect classes to 17, by collecting data from 17 Arabic speaking countries. Both tasks use YouTube recordings to provide a multi-genre multi-dialectal challenge in the wild. Moroccan speech transcription used about 13 hours of transcribed speech data, split across training, development, and test sets, covering 7-genres: comedy, cooking, family/kids, fashion, drama, sports, and science (TEDx). The fine-grained Arabic dialect identification data was collected from known YouTube channels from 17 Arabic countries. 3,000 hours of this data was released for training, and 57 hours for development and testing. The dialect identification data was divided into three sub-categories based on the segment duration: short (under 5 s), medium (5–20 s), and long (>20 s). Overall, 25 teams registered for the challenge, and 9 teams submitted systems for the two tasks. We outline the approaches adopted in each system and summarize the evaluation results. | nan |
TSAC | [] | https://huggingface.co/datasets/arbml/TSAC | https://github.com/fbougares/TSAC | LGPL-3.0 | 2,017 | ar | ar-TN: (Arabic (Tunisia)) | social media | text | crawling and annotation(other) | About 17k user comments manually annotated to positive and negative polarities. This corpus is collected from Facebook users comments written on official pages of Tunisian radios and TV channels | 17,000 | sentences | Medium | Vienna | nan | Sentiment Analysis of Tunisian Dialect:
Linguistic Resources and Experiments | https://aclanthology.org/W17-1307.pdf | Arab | No | GitHub | Free | nan | Yes | sentiment analysis | WANLP | 59.0 | workshop | Arabic Natural Language Processing Workshop | Salima Medhaffar,Fethi Bougares,Y. Estève,L. Belguith | ,,, | Dialectal Arabic (DA) is significantly different from the Arabic language taught in schools and used in written communication and formal speech (broadcast news, religion, politics, etc.). There are many existing researches in the field of Arabic language Sentiment Analysis (SA); however, they are generally restricted to Modern Standard Arabic (MSA) or some dialects of economic or political interest. In this paper we are interested in the SA of the Tunisian Dialect. We utilize Machine Learning techniques to determine the polarity of comments written in Tunisian Dialect. First, we evaluate the SA systems performances with models trained using freely available MSA and Multi-dialectal data sets. We then collect and annotate a Tunisian Dialect corpus of 17.000 comments from Facebook. This corpus allows us a significant accuracy improvement compared to the best model trained on other Arabic dialects or MSA data. We believe that this first freely available corpus will be valuable to researchers working in the field of Tunisian Sentiment Analysis and similar areas. | nan |
NileULex | [] | https://huggingface.co/datasets/arbml/NileULex | https://github.com/NileTMRG/NileULex | custom | 2,016 | ar | mixed | social media | text | crawling and annotation(other) | Egyptian Arabic and Modern Standard Arabic sentiment words and their polarity | 5,953 | sentences | Medium | Nile University | nan | NileULex: A Phrase and Word Level Sentiment Lexicon for Egyptian and
Modern Standard Arabic | https://aclanthology.org/L16-1463.pdf | Arab | No | GitHub | Free | nan | No | sentiment analysis | LREC | 39.0 | conference | International Conference on Language Resources and Evaluation | S. El-Beltagy | nan | This paper presents NileULex, which is an Arabic sentiment lexicon containing close to six thousands Arabic words and compound phrases. Forty five percent of the terms and expressions in the lexicon are Egyptian or colloquial while fifty five percent are Modern Standard Arabic. While the collection of many of the terms included in the lexicon was done automatically, the actual addition of any term was done manually. One of the important criterions for adding terms to the lexicon, was that they be as unambiguous as possible. The result is a lexicon with a much higher quality than any translated variant or automatically constructed one. To demonstrate that a lexicon such as this can directly impact the task of sentiment analysis, a very basic machine learning based sentiment analyser that uses unigrams, bigrams, and lexicon based features was applied on two different Twitter datasets. The obtained results were compared to a baseline system that only uses unigrams and bigrams. The same lexicon based features were also generated using a publicly available translation of a popular sentiment lexicon. The experiments show that usage of the developed lexicon improves the results over both the baseline and the publicly available lexicon. | nan |
CALLHOME: Egyptian Arabic Speech Translation Corpus | [] | nan | https://github.com/noisychannel/ARZ_callhome_corpus | CC BY-SA 4.0 | 2,014 | multilingual | ar-EG: (Arabic (Egypt)) | social media | text | human translation | three-way parallel dataset of Egyptian Arabic Speech, transcriptions and English translations | 39,213 | sentences | Medium | Multiple institutions | nan | TRANSLATIONS OF THE CALLHOME EGYPTIAN ARABIC CORPUS FOR
CONVERSATIONAL SPEECH TRANSLATION | https://www.cis.upenn.edu/~ccb/publications/callhome-egyptian-arabic-speech-translations.pdf | Arab | No | GitHub | Free | nan | Yes | machine translation | other | 10.0 | preprint | nan | G. Kumar,Yuan Cao,Ryan Cotterell,Chris Callison-Burch,Daniel Povey,S. Khudanpur | ,Google Brain,,,, | Translation of the output of automatic speech recognition (ASR) systems, also known as speech translation, has received a lot of research interest recently. This is especially true for programs such as DARPA BOLT which focus on improving spontaneous human-human conversation across languages. However, this research is hindered by the dearth of datasets developed for this explicit purpose. For Egyptian Arabic-English, in particular, no parallel speechtranscription-translation dataset exists in the same domain. In order to support research in speech translation, we introduce the Callhome Egyptian Arabic-English Speech Translation Corpus. This supplements the existing LDC corpus with four reference translations for each utterance in the transcripts. The result is a three-way parallel dataset of Egyptian Arabic Speech, transcriptions and English translations. | nan |
Comparable Wikipedia Coprus | [
{
"Name": "Arabic Wikipedia",
"Dialect": "ar-MSA: (Arabic (Modern Standard Arabic))",
"Volume": "10,197",
"Unit": "documents"
},
{
"Name": "Egyptian Wikipedia",
"Dialect": "ar-EG: (Arabic (Egypt))",
"Volume": "10,197",
"Unit": "documents"
}
] | nan | https://github.com/motazsaad/comparableWikiCoprus | CC BY-SA 4.0 | 2,017 | ar | mixed | wikipedia | text | crawling and annotation(other) | Comparable Wikipedia Corpus (aligned documents) Corpus extracts from 20-01-2017 Wikipedia dumps | 20,394 | documents | Low | Islamic University of Gaza | nan | WikiDocsAligner: An Off-the-Shelf Wikipedia Documents Alignment Tool | https://ieeexplore.ieee.org/document/8038320 | Arab | No | GitHub | Free | nan | No | machine translation | PICICT | 4.0 | conference | Palestinian International Conference on Information and Communication Technology | Motaz Saad,B. Alijla | The Islamic University of Gaza, | Wikipedia encyclopedia is an attractive source for comparable corpora in many languages. Most researchers develop their own script to perform document alignment task, which requires efforts and time. In this paper, we present WikiDocsAligner, an off-the-shelf Wikipedia Articles alignment handy tool. The implementation of WikiDocsAligner does not require the researchers to import/export of interlanguage links databases. The user just need to download Wikipedia dumps (interlanguage links and articles), then provide them to the tool, which performs the alignment. This software can be used easily to align Wikipedia documents in any language pair. Finally, we use WikiDocsAligner to align comparable documents from Arabic Wikipedia and Egyptian Wikipedia. So we shed the light on Wikipedia as a source of Arabic dialects language resources. The produced resources is interesting and useful as the demand on Arabic/dialects language resources increased in the last decade. | nan |
AOC | [
{
"Name": "MSA",
"Dialect": "ar-MSA: (Arabic (Modern Standard Arabic))",
"Volume": "63,555",
"Unit": "sentences"
},
{
"Name": "Dialectal",
"Dialect": "mixed",
"Volume": "44,618",
"Unit": "sentences"
}
] | nan | https://github.com/sjeblee/AOC | unknown | 2,011 | ar | mixed | news articles | text | crawling and annotation(other) | a 52M-word monolingual dataset rich in dialectal content | 108,000 | sentences | Low | Johns Hopkins University | nan | The Arabic Online Commentary Dataset: an Annotated Dataset of Informal Arabic with High Dialectal Content | https://aclanthology.org/P11-2007.pdf | Arab | No | GitHub | Free | nan | No | dialect identification | ACL | 147.0 | conference | Assofications of computation linguisitcs | Omar Zaidan,Chris Callison-Burch | , | The written form of Arabic, Modern Standard Arabic (MSA), differs quite a bit from the spoken dialects of Arabic, which are the true "native" languages of Arabic speakers used in daily life. However, due to MSA's prevalence in written form, almost all Arabic datasets have predominantly MSA content. We present the Arabic Online Commentary Dataset, a 52M-word monolingual dataset rich in dialectal content, and we describe our long-term annotation effort to identify the dialect level (and dialect itself) in each sentence of the dataset. So far, we have labeled 108K sentences, 41% of which as having dialectal content. We also present experimental results on the task of automatic dialect identification, using the collected labels for training and evaluation. | nan |
PADIC | [
{
"Name": "ALG",
"Dialect": "ar-DZ: (Arabic (Algeria))",
"Volume": "6,400",
"Unit": "sentences"
},
{
"Name": "ANB",
"Dialect": "ar-DZ: (Arabic (Algeria))",
"Volume": "6,400",
"Unit": "sentences"
},
{
"Name": "TUN",
"Dialect": "ar-TN: (Arabic (Tunisia))",
"Volume": "6,400",
"Unit": "sentences"
},
{
"Name": "SYR",
"Dialect": "ar-SY: (Arabic (Syria))",
"Volume": "6,400",
"Unit": "sentences"
},
{
"Name": "PAL",
"Dialect": "ar-PS: (Arabic (Palestine))",
"Volume": "6,400",
"Unit": "sentences"
},
{
"Name": "MSA",
"Dialect": "ar-MSA: (Arabic (Modern Standard Arabic))",
"Volume": "6,400",
"Unit": "sentences"
}
] | https://huggingface.co/datasets/arbml/PADIC | https://sourceforge.net/projects/padic/ | GPL-3.0 | 2,015 | ar | mixed | other | text | manual curation | a Parallel Arabic DIalect Corpus we built from scratch, | 32,000 | sentences | Low | Multiple institutions | nan | Machine Translation Experiments on PADIC:
A Parallel Arabic DIalect Corpus | https://aclanthology.org/Y15-1004.pdf | Arab-Latn | No | other | Free | nan | No | machine translation | PACLIC | 58.0 | conference | Pacific Asia Conference on Language, Information and Computation | Karima Meftouh,S. Harrat,S. Jamoussi,Mourad Abbas,Kamel Smaïli | ,,,, | We present in this paper PADIC, a Parallel Arabic DIalect Corpus we built from scratch, then we conducted experiments on crossdialect Arabic machine translation. PADIC is composed of dialects from both the Maghreb and the Middle-East. Each dialect has been aligned with Modern Standard Arabic (MSA). Three dialects from Maghreb are concerned by this study: two from Algeria, one from Tunisia, and two dialects from the MiddleEast (Syria and Palestine). PADIC has been built from scratch because the lack of dialect resources. In fact, Arabic dialects in Arab world in general are used in daily life conversations but they are not written. At the best of our knowledge, PADIC, up to now, is the largest corpus in the community working on dialects and especially those concerning Maghreb. PADIC is composed of 6400 sentences for each of the 5 concerned dialects and MSA. We conducted cross-lingual machine translation experiments between all the language pairs. For translating to MSA we interpolated the corresponding Language Model (LM) with a large Arabic corpus based LM. We also studied the impact of language model smoothing techniques on the results of machine translation because this corpus, even it is the largest one, it still very small in comparison to those used for translation of natural languages. | nan |
Habibi | [
{
"Name": "Gulf",
"Dialect": "ar-GLF: (Arabic (Gulf))",
"Volume": "9,484",
"Unit": "documents"
},
{
"Name": "Egyptian",
"Dialect": "ar-EG: (Arabic (Egypt))",
"Volume": "7,265",
"Unit": "documents"
},
{
"Name": "Levantine",
"Dialect": "ar-LEV: (Arabic(Levant))",
"Volume": "6,016",
"Unit": "documents"
},
{
"Name": "Iraqi",
"Dialect": "ar-IQ: (Arabic (Iraq))",
"Volume": "3,438",
"Unit": "documents"
},
{
"Name": "Sudan",
"Dialect": "ar-SD: (Arabic (Sudan))",
"Volume": "2,662",
"Unit": "documents"
},
{
"Name": "Maghrebi",
"Dialect": "ar-NOR: (Arabic (North Africa))",
"Volume": "1,207",
"Unit": "documents"
}
] | https://huggingface.co/datasets/arbml/Habibi | https://www.lancaster.ac.uk/staff/elhaj/corpora.html | unknown | 2,020 | ar | mixed | other | text | crawling | The corpus comprises more than 30,000 Arabic song lyrics in 6 Arabic dialects for singers from 18 different Arabic countries. | 30,072 | documents | Low | Lancaster University | nan | Habibi - a multi Dialect multi National Arabic Song Lyrics Corpus
| https://aclanthology.org/2020.lrec-1.165.pdf | Arab | No | other | Free | nan | No | text generation, language modeling | LREC | 7.0 | conference | International Conference on Language Resources and Evaluation | Mahmoud El-Haj | nan | This paper introduces Habibi the first Arabic Song Lyrics corpus. The corpus comprises more than 30,000 Arabic song lyrics in 6 Arabic dialects for singers from 18 different Arabic countries. The lyrics are segmented into more than 500,000 sentences (song verses) with more than 3.5 million words. I provide the corpus in both comma separated value (csv) and annotated plain text (txt) file formats. In addition, I converted the csv version into JavaScript Object Notation (json) and eXtensible Markup Language (xml) file formats. To experiment with the corpus I run extensive binary and multi-class experiments for dialect and country-of-origin identification. The identification tasks include the use of several classical machine learning and deep learning models utilising different word embeddings. For the binary dialect identification task the best performing classifier achieved a testing accuracy of 93%. This was achieved using a word-based Convolutional Neural Network (CNN) utilising a Continuous Bag of Words (CBOW) word embeddings model. The results overall show all classical and deep learning models to outperform our baseline, which demonstrates the suitability of the corpus for both dialect and country-of-origin identification tasks. I am making the corpus and the trained CBOW word embeddings freely available for research purposes. | nan |
KALIMAT | [] | nan | https://sourceforge.net/projects/kalimat/ | custom | 2,013 | ar | ar-MSA: (Arabic (Modern Standard Arabic)) | news articles | text | crawling | 20,291 Arabic articles collected from the Omani newspaper Alwatan | 20,291 | documents | Low | Multiple institutions | nan | KALIMAT a Multipurpose Arabic
Corpus | https://eprints.lancs.ac.uk/id/eprint/71282/1/KALIMAT_ELHAJ_KOULALI.pdf | Arab | No | sourceforge | Free | nan | Yes | topic classification,summarization,named entity recognition,part of speech tagging,morphological analysis | other | 30.0 | preprint | nan | Mahmoud El-Haj,R. Koulali | Lancaster University, | Resources, such as corpora, are important for researchers working on Arabic Natural Language Processing (NLP) (Al-Sulaiti et al. 2006). For this reason we came up with the idea of generating an Arabic multipurpose corpus, which we call KALIMAT (Arabic transliteration of “WORDS”). The automatically created corpus could benefit researchers working on different Arabic NLP areas. In our work on Arabic we developed, enhanced and tested many Arabic NLP tools. We tuned these tools to provide high quality results. The tools include auto-summarisers, Part of Speech Tagger, Morphological Analyser and Named Entity Recognition (NER). We ran these tools using the same document collection. We provide the output corpus freely for researchers to evaluate their work and to run experiments for different Arabic NLP purposes using one corpus. | nan |
EASC | [] | https://huggingface.co/datasets/arbml/EASC | https://sourceforge.net/projects/easc-corpus/ | CC BY-SA 3.0 | 2,010 | ar | ar-MSA: (Arabic (Modern Standard Arabic)) | news articles | text | crawling | 153 Arabic articles and 765 human-generated extractive summaries of those articles | 153 | documents | Low | University of Essex | nan | Using Mechanical Turk to Create a Corpus of Arabic Summaries
| http://repository.essex.ac.uk/4064/1/LREC2010_MTurk.pdf | Arab | No | sourceforge | Free | nan | No | summarization | other | 59.0 | preprint | nan | Mahmoud El-Haj,Udo Kruschwitz,C. Fox | Lancaster University,University of Regensburg, | This paper describes the creation of a human-generated corpus of extractive Arabic summaries of a selection of Wikipedia and Arabic newspaper articles using Mechanical Turk?an online workforce. The purpose of this exercise was two-fold. First, it addresses a shortage of relevant data for Arabic natural language processing. Second, it demonstrates the application of Mechanical Turk to the problem of creating natural language resources. The paper also reports on a number of evaluations we have performed to compare the collected summaries against results obtained from a variety of automatic summarisation systems. | nan |
Arabic Dialects Dataset | [
{
"Name": "GLF",
"Dialect": "ar-GLF: (Arabic (Gulf))",
"Volume": "2,546",
"Unit": "sentences"
},
{
"Name": "LAV",
"Dialect": "ar-LEV: (Arabic(Levant))",
"Volume": "2,463",
"Unit": "sentences"
},
{
"Name": "MSA",
"Dialect": "ar-MSA: (Arabic (Modern Standard Arabic))",
"Volume": "3,731",
"Unit": "sentences"
},
{
"Name": "NOR",
"Dialect": "ar-NOR: (Arabic (North Africa))",
"Volume": "3,693",
"Unit": "sentences"
},
{
"Name": "EGY",
"Dialect": "ar-EG: (Arabic (Egypt))",
"Volume": "4,061",
"Unit": "sentences"
}
] | https://huggingface.co/datasets/arbml/Arabic_Dialects_Dataset | https://www.lancaster.ac.uk/staff/elhaj/corpora.html | unknown | 2,018 | ar | mixed | other | text | crawling and annotation(other) | Dataset of Arabic dialects for GULF, EGYPT, LEVANT, TONESIAN Arabic dialects in addition to MSA. | 16,494 | sentences | Low | Lancaster University | OAC | Arabic Dialect Identification in the Context of Bivalency and Code-Switching
| https://aclanthology.org/L18-1573.pdf | Arab | No | other | Free | nan | No | dialect identification | LREC | 17.0 | conference | International Conference on Language Resources and Evaluation | Mahmoud El-Haj,Paul Rayson,Mariam Aboelezz | Lancaster University,Lancaster University, | In this paper we use a novel approach towards Arabic dialect identification using language bivalency and written code-switching. Bivalency between languages or dialects is where a word or element is treated by language users as having a fundamentally similar semantic content in more than one language or dialect. Arabic dialect identification in writing is a difficult task even for humans due to the fact that words are used interchangeably between dialects. The task of automatically identifying dialect is harder and classifiers trained using only n-grams will perform poorly when tested on unseen data. Such approaches require significant amounts of annotated training data which is costly and time consuming to produce. Currently available Arabic dialect datasets do not exceed a few hundred thousand sentences, thus we need to extract features other than word and character n-grams. In our work we present experimental results from automatically identifying dialects from the four main Arabic dialect regions (Egypt, North Africa, Gulf and Levant) in addition to Standard Arabic. We extend previous work by incorporating additional grammatical and stylistic features and define a subtractive bivalency profiling approach to address issues of bivalent words across the examined Arabic dialects. The results show that our new methods classification accuracy can reach more than 76% and score well (66%) when tested on completely unseen data. | nan |
ANERcorp | [] | nan | https://camel.abudhabi.nyu.edu/anercorp/ | CC BY-SA 4.0 | 2,020 | ar | ar-MSA: (Arabic (Modern Standard Arabic)) | news articles | text | crawling and annotation(other) | collected from different resources | 316 | documents | Low | NYU Abu Dhabi University | nan | CAMeL Tools: An Open Source Python Toolkit for Arabic Natural Language Processing | https://aclanthology.org/2020.lrec-1.868.pdf | Arab | No | CAMeL Resources | Upon-Request | nan | Yes | named entity recognition | LREC | 22.0 | conference | International Conference on Language Resources and Evaluation | Ossama Obeid,Nasser Zalmout,Salam Khalifa,Dima Taji,M. Oudah,Bashar Alhafni,Go Inoue,Fadhl Eryani,Alexander Erdmann,Nizar Habash | ,,New York University Abu Dhabi,,,,New York University;New York University Abu Dhabi,,, | We present CAMeL Tools, a collection of open-source tools for Arabic natural language processing in Python. CAMeL Tools currently provides utilities for pre-processing, morphological modeling, Dialect Identification, Named Entity Recognition and Sentiment Analysis. In this paper, we describe the design of CAMeL Tools and the functionalities it provides. | nan |
APGC v1.0: Arabic Parallel Gender Corpus v1.0 | [] | nan | https://camel.abudhabi.nyu.edu/arabic-parallel-gender-corpus/ | custom | 2,019 | multilingual | ar-MSA: (Arabic (Modern Standard Arabic)) | other | text | crawling and annotation(other) | a corpus designed to support research on gender bias in natural language processing applications working on Arabic | 12,000 | sentences | Low | Multiple institutions | OpenSubtitles | Automatic Gender Identification and Reinflection in Arabic
| https://aclanthology.org/W19-3822v2.pdf | Arab | No | CAMeL Resources | Upon-Request | nan | Yes | gender identification, gender rewriting | GeBNLP | 13.0 | workshop | Workshop on Gender Bias in Natural Language Processing | Nizar Habash,Houda Bouamor,Christine Chung | ,, | The impressive progress in many Natural Language Processing (NLP) applications has increased the awareness of some of the biases these NLP systems have with regards to gender identities. In this paper, we propose an approach to extend biased single-output genderblind NLP systems with gender-specific alternative reinflections. We focus on Arabic, a gender-marking morphologically rich language, in the context of machine translation (MT) from English, and for first-personsingular constructions only. Our contributions are the development of a system-independent gender-awareness wrapper, and the building of a corpus for training and evaluating firstperson-singular gender identification and reinflection in Arabic. Our results successfully demonstrate the viability of this approach with 8% relative increase in BLEU score for firstperson-singular feminine, and 5.3% comparable increase for first-person-singular masculine on top of a state-of-the-art gender-blind MT system on a held-out test set. | nan |
TUFS Media | [] | nan | http://ngc2068.tufs.ac.jp/tufsmedia-corpus/ | unknown | 2,018 | multilingual | ar-MSA: (Arabic (Modern Standard Arabic)) | news articles | text | crawling and annotation(translation) | a parallel corpus of translated news articles collected at Tokyo University of Foreign Studies (TUFS) | 8,652 | sentences | Low | Tokyo University | nan | A Parallel Corpus of Arabic–Japanese News Articles
| https://aclanthology.org/L18-1147.pdf | Arab | No | other | Upon-Request | nan | Yes | machine translation | LREC | 9.0 | conference | International Conference on Language Resources and Evaluation | Go Inoue,Nizar Habash,Yuji Matsumoto,Hiroyuki Aoyama | New York University;New York University Abu Dhabi,,, | Much work has been done on machine translation between major language pairs including Arabic–English and English–Japanese thanks to the availability of large-scale parallel corpora with manually verified subsets of parallel sentences. However, there has been little research conducted on the Arabic–Japanese language pair due to its parallel-data scarcity, despite being a good example of interestingly contrasting differences in typology. In this paper, we describe the creation process and statistics of the Arabic–Japanese portion of the TUFS Media Corpus, a parallel corpus of translated news articles collected at Tokyo University of Foreign Studies (TUFS). Part of the corpus is manually aligned at the sentence level for development and testing. The corpus is provided in two formats: A document-level parallel corpus in XML format, and a sentence-level parallel corpus in plain text format. We also report the first results of Arabic– Japanese phrase-based machine translation trained on our corpus. | nan |
United Nations Parallel Corpus | [] | https://huggingface.co/datasets/un_pc | https://conferences.unite.un.org/uncorpus | custom | 2,016 | multilingual | ar-MSA: (Arabic (Modern Standard Arabic)) | other | text | human translation | The parallel corpus presented consists of manually translated UN documents from the last 25 years | 540,152 | documents | Low | United Nations | nan | The United Nations Parallel Corpus v1.0 | https://conferences.unite.un.org/UNCORPUS/Content/Doc/un.pdf | Arab | No | other | Free | nan | Yes | machine translation | LREC | 233.0 | conference | International Conference on Language Resources and Evaluation | Michal Ziemski,Marcin Junczys-Dowmunt,B. Pouliquen | ,, | This paper describes the creation process and statistics of the official United Nations Parallel Corpus, the first parallel corpus composed from United Nations documents published by the original data creator. The parallel corpus presented consists of manually translated UN documents from the last 25 years (1990 to 2014) for the six official UN languages, Arabic, Chinese, English, French, Russian, and Spanish. The corpus is freely available for download under a liberal license. Apart from the pairwise aligned documents, a fully aligned subcorpus for the six official UN languages is distributed. We provide baseline BLEU scores of our Moses-based SMT systems trained with the full data of language pairs involving English and for all possible translation directions of the six-way subcorpus. | nan |
NADI-2020 | [
{
"Name": "Algeria",
"Dialect": "ar-DZ: (Arabic (Algeria))",
"Volume": "2,214",
"Unit": "sentences"
},
{
"Name": "Bahrain",
"Dialect": "ar-BH: (Arabic (Bahrain))",
"Volume": "238",
"Unit": "sentences"
},
{
"Name": "Djibouti",
"Dialect": "ar-DJ: (Arabic (Djibouti))",
"Volume": "271",
"Unit": "sentences"
},
{
"Name": "Egypt",
"Dialect": "ar-EG: (Arabic (Egypt))",
"Volume": "6,635",
"Unit": "sentences"
},
{
"Name": "Iraq",
"Dialect": "ar-IQ: (Arabic (Iraq))",
"Volume": "3,816",
"Unit": "sentences"
},
{
"Name": "Jordan",
"Dialect": "ar-JO: (Arabic (Jordan))",
"Volume": "634",
"Unit": "sentences"
},
{
"Name": "Kuwait",
"Dialect": "ar-KW: (Arabic (Kuwait))",
"Volume": "592",
"Unit": "sentences"
},
{
"Name": "Lebanon",
"Dialect": "ar-LB: (Arabic (Lebanon))",
"Volume": "905",
"Unit": "sentences"
},
{
"Name": "Libya",
"Dialect": "ar-LY: (Arabic (Libya))",
"Volume": "1,600",
"Unit": "sentences"
},
{
"Name": "Mauritania",
"Dialect": "ar-MR: (Arabic (Mauritania))",
"Volume": "255",
"Unit": "sentences"
},
{
"Name": "Morocco",
"Dialect": "ar-MA: (Arabic (Morocco))",
"Volume": "1,579",
"Unit": "sentences"
},
{
"Name": "Oman",
"Dialect": "ar-OM: (Arabic (Oman))",
"Volume": "1,615",
"Unit": "sentences"
},
{
"Name": "Palestine",
"Dialect": "ar-PS: (Arabic (Palestine))",
"Volume": "624",
"Unit": "sentences"
},
{
"Name": "Qatar ",
"Dialect": "ar-QA: (Arabic (Qatar))",
"Volume": "399",
"Unit": "sentences"
},
{
"Name": "Saudi Arabia",
"Dialect": "ar-SA: (Arabic (Saudi Arabia))",
"Volume": "3,455",
"Unit": "sentences"
},
{
"Name": "Somalia",
"Dialect": "ar-SO: (Arabic (Somalia))",
"Volume": "312",
"Unit": "sentences"
},
{
"Name": "Sudan",
"Dialect": "ar-SD: (Arabic (Sudan))",
"Volume": "312",
"Unit": "sentences"
},
{
"Name": "Syria",
"Dialect": "ar-SY: (Arabic (Syria))",
"Volume": "1,595",
"Unit": "sentences"
},
{
"Name": "Tunisia",
"Dialect": "ar-TN: (Arabic (Tunisia))",
"Volume": "1,122",
"Unit": "sentences"
},
{
"Name": "UAE",
"Dialect": "ar-AE: (Arabic (United Arab Emirates))",
"Volume": "1,548",
"Unit": "sentences"
},
{
"Name": "Yemen",
"Dialect": "ar-YE: (Arabic (Yemen))",
"Volume": "1,236",
"Unit": "sentences"
}
] | nan | https://sites.google.com/view/nadi-shared-task | custom | 2,020 | ar | mixed | social media | text | crawling and annotation(other) | The data for the shared task covers a total of 100 provinces from 21 Arab countries and are collected from the Twitter domain | 30,957 | sentences | Medium | Multiple institutions | nan | NADI 2020: The First Nuanced Arabic Dialect Identification Shared Task
| https://arxiv.org/pdf/2010.11334.pdf | Arab | No | other | Upon-Request | nan | Yes | dialect identification | WANLP | 38.0 | workshop | Arabic Natural Language Processing Workshop | Muhammad Abdul-Mageed,Chiyu Zhang,Houda Bouamor,Nizar Habash | ,The University of British Columbia,, | We present the results and findings of the First Nuanced Arabic Dialect Identification Shared Task (NADI). The shared task includes two subtasks: country level dialect identification (Subtask 1) and province level sub-dialect identification (Subtask 2). The data for the shared task covers a total of 100 provinces from 21 Arab countries, and are collected from the Twitter domain. As such, NADI is the first shared task to target naturally-occurring fine-grained dialectal text at the sub-country level. A total of 61 teams from 25 countries registered to participate in the tasks, thus reflecting the interest of the community in this area. We received 47 submissions for Subtask 1 from 18 teams and 9 submissions to Subtask 2 from 9 teams. | nan |
NADI-2021 | [
{
"Name": "Algeria",
"Dialect": "ar-DZ: (Arabic (Algeria))",
"Volume": "2,765",
"Unit": "sentences"
},
{
"Name": "Bahrain",
"Dialect": "ar-BH: (Arabic (Bahrain))",
"Volume": "313",
"Unit": "sentences"
},
{
"Name": "Djibouti",
"Dialect": "ar-DJ: (Arabic (Djibouti))",
"Volume": "314",
"Unit": "sentences"
},
{
"Name": "Egypt",
"Dialect": "ar-EG: (Arabic (Egypt))",
"Volume": "6,241",
"Unit": "sentences"
},
{
"Name": "Iraq",
"Dialect": "ar-IQ: (Arabic (Iraq))",
"Volume": "4,042",
"Unit": "sentences"
},
{
"Name": "Jordan",
"Dialect": "ar-JO: (Arabic (Jordan))",
"Volume": "627",
"Unit": "sentences"
},
{
"Name": "Kuwait",
"Dialect": "ar-KW: (Arabic (Kuwait))",
"Volume": "627",
"Unit": "sentences"
},
{
"Name": "Lebanon",
"Dialect": "ar-LB: (Arabic (Lebanon))",
"Volume": "929",
"Unit": "sentences"
},
{
"Name": "Libya",
"Dialect": "ar-LY: (Arabic (Libya))",
"Volume": "1,883",
"Unit": "sentences"
},
{
"Name": "Mauritania",
"Dialect": "ar-MR: (Arabic (Mauritania))",
"Volume": "314",
"Unit": "sentences"
},
{
"Name": "Morocco",
"Dialect": "ar-MA: (Arabic (Morocco))",
"Volume": "1,256",
"Unit": "sentences"
},
{
"Name": "Oman",
"Dialect": "ar-OM: (Arabic (Oman))",
"Volume": "2,175",
"Unit": "sentences"
},
{
"Name": "Palestine",
"Dialect": "ar-PS: (Arabic (Palestine))",
"Volume": "626",
"Unit": "sentences"
},
{
"Name": "Qatar ",
"Dialect": "ar-QA: (Arabic (Qatar))",
"Volume": "314",
"Unit": "sentences"
},
{
"Name": "Saudi Arabia",
"Dialect": "ar-SA: (Arabic (Saudi Arabia))",
"Volume": "3,130",
"Unit": "sentences"
},
{
"Name": "Somalia",
"Dialect": "ar-SO: (Arabic (Somalia))",
"Volume": "511",
"Unit": "sentences"
},
{
"Name": "Sudan",
"Dialect": "ar-SD: (Arabic (Sudan))",
"Volume": "310",
"Unit": "sentences"
},
{
"Name": "Syria",
"Dialect": "ar-SY: (Arabic (Syria))",
"Volume": "1,881",
"Unit": "sentences"
},
{
"Name": "Tunisia",
"Dialect": "ar-TN: (Arabic (Tunisia))",
"Volume": "1,190",
"Unit": "sentences"
},
{
"Name": "UAE",
"Dialect": "ar-AE: (Arabic (United Arab Emirates))",
"Volume": "940",
"Unit": "sentences"
},
{
"Name": "Yemen",
"Dialect": "ar-YE: (Arabic (Yemen))",
"Volume": "612",
"Unit": "sentences"
}
] | nan | https://sites.google.com/view/nadi-shared-task | CC BY-NC-ND 4.0 | 2,021 | ar | mixed | social media | text | crawling and annotation(other) | The shared task dataset covers a total of 100 provinces from 21 Arab countries, collected from the Twitter domain. | 310,000 | sentences | Medium | Multiple institutions | nan | NADI 2021:
The Second Nuanced Arabic Dialect Identification Shared Task
| https://arxiv.org/pdf/2103.08466.pdf | Arab | No | other | Upon-Request | nan | Yes | dialect identification | WANLP | 12.0 | workshop | Arabic Natural Language Processing Workshop | Muhammad Abdul-Mageed,Chiyu Zhang,AbdelRahim Elmadany,Houda Bouamor,Nizar Habash | ,The University of British Columbia,University of British Columbia,, | We present the findings and results of theSecond Nuanced Arabic Dialect IdentificationShared Task (NADI 2021). This Shared Taskincludes four subtasks: country-level ModernStandard Arabic (MSA) identification (Subtask1.1), country-level dialect identification (Subtask1.2), province-level MSA identification (Subtask2.1), and province-level sub-dialect identifica-tion (Subtask 2.2). The shared task dataset cov-ers a total of 100 provinces from 21 Arab coun-tries, collected from the Twitter domain. A totalof 53 teams from 23 countries registered to par-ticipate in the tasks, thus reflecting the interestof the community in this area. We received 16submissions for Subtask 1.1 from five teams, 27submissions for Subtask 1.2 from eight teams,12 submissions for Subtask 2.1 from four teams,and 13 Submissions for subtask 2.2 from fourteams. | nan |
AraStance | [] | https://huggingface.co/datasets/arbml/arastance | https://github.com/Tariq60/arastance | unknown | 2,021 | ar | ar-MSA: (Arabic (Modern Standard Arabic)) | news articles | text | crawling and annotation(other) | covers false and true claims from multiple domains (e.g., politics, sports, health) and several Arab countries | 4,063 | sentences | Low | Multiple institutions | nan | AraStance: A Multi-Country and Multi-Domain Dataset of
Arabic Stance Detection for Fact Checking | https://arxiv.org/pdf/2104.13559.pdf | Arab | No | GitHub | Free | nan | Yes | stance detection | NLP4IF | 0.0 | workshop | NLP for Internet Freedom | Tariq Alhindi,Amal Alabdulkarim,A. Alshehri,Muhammad Abdul-Mageed,Preslav Nakov | Columbia University;King Abdulaziz City for Science and Technology,,,, | With the continuing spread of misinformation and disinformation online, it is of increasing importance to develop combating mechanisms at scale in the form of automated systems that support multiple languages. One task of interest is claim veracity prediction, which can be addressed using stance detection with respect to relevant documents retrieved online. To this end, we present our new Arabic Stance Detection dataset (AraStance) of 4,063 claim–article pairs from a diverse set of sources comprising three fact-checking websites and one news website. AraStance covers false and true claims from multiple domains (e.g., politics, sports, health) and several Arab countries, and it is well-balanced between related and unrelated documents with respect to the claims. We benchmark AraStance, along with two other stance detection datasets, using a number of BERT-based models. Our best model achieves an accuracy of 85% and a macro F1 score of 78%, which leaves room for improvement and reflects the challenging nature of AraStance and the task of stance detection in general. | nan |
QCRI Parallel Tweets | [] | https://huggingface.co/datasets/tweets_ar_en_parallel | https://alt.qcri.org/resources/bilingual_corpus_of_parallel_tweets | Apache-2.0 | 2,020 | multilingual | mixed | social media | text | crawling | bilingual corpus of Arabic-English parallel tweets and a list of Twitter accounts who post Arabic-English | 166,000 | sentences | Medium | QCRI | nan | Constructing a Bilingual Corpus of Parallel Tweets
| https://aclanthology.org/2020.bucc-1.3.pdf | Arab | No | QCRI Resources | Free | nan | No | machine translation | BUCC | 3.0 | workshop | Workshop on Building and Using Comparable Corpora | Hamdy Mubarak,Sabit Hassan,Ahmed Abdelali | ,, | In a bid to reach a larger and more diverse audience, Twitter users often post parallel tweets—tweets that contain the same content but are written in different languages. Parallel tweets can be an important resource for developing machine translation (MT) systems among other natural language processing (NLP) tasks. In this paper, we introduce a generic method for collecting parallel tweets. Using this method, we collect a bilingual corpus of English-Arabic parallel tweets and a list of Twitter accounts who post English-Arabictweets regularly. Since our method is generic, it can also be used for collecting parallel tweets that cover less-resourced languages such as Serbian and Urdu. Additionally, we annotate a subset of Twitter accounts with their countries of origin and topic of interest, which provides insights about the population who post parallel tweets. This latter information can also be useful for author profiling tasks. | nan |
Arabic ALA LC Romanization | [] | https://huggingface.co/datasets/arbml/ALA_LC_Romanization | https://github.com/CAMeL-Lab/Arabic_ALA-LC_Romanization | unknown | 2,021 | ar | ar-MSA: (Arabic (Modern Standard Arabic)) | other | text | crawling | parallel Arabic and Romanized bibliographic entries | 107,439 | sentences | Low | NYU Abu Dhabi | nan | Automatic Romanization of Arabic Bibliographic Records
| https://aclanthology.org/2021.wanlp-1.23.pdf | Arab-Latn | No | GitHub | Free | nan | Yes | text romanization | WANLP | 0.0 | workshop | Arabic Natural Language Processing Workshop | Fadhl Eryani,Nizar Habash | , | International library standards require cataloguers to tediously input Romanization of their catalogue records for the benefit of library users without specific language expertise. In this paper, we present the first reported results on the task of automatic Romanization of undiacritized Arabic bibliographic entries. This complex task requires the modeling of Arabic phonology, morphology, and even semantics. We collected a 2.5M word corpus of parallel Arabic and Romanized bibliographic entries, and benchmarked a number of models that vary in terms of complexity and resource dependence. Our best system reaches 89.3% exact word Romanization on a blind test set. We make our data and code publicly available. | nan |
TALAA | [] | https://huggingface.co/datasets/arbml/TALAA | https://github.com/saidziani/Arabic-News-Article-Classification | unknown | 2,015 | ar | ar-MSA: (Arabic (Modern Standard Arabic)) | news articles | text | crawling | collections of articles | 57,827 | documents | Low | USTHB Algeria | nan | Building TALAA, a Free General and Categorized Arabic Corpus
| https://www.scitepress.org/Papers/2015/53521/53521.pdf | Arab | No | GitHub | Free | nan | Yes | topic classification | ICAART | 4.0 | conference | Conference on Agents and Artificial Intelligence | Essma Selab,A. Guessoum | , | Arabic natural language processing (ANLP) has gained increasing interest over the last decade. However,
the development of ANLP tools depends on the availability of large corpora. It turns out unfortunately that
the scientific community has a deficit in large and varied Arabic corpora, especially ones that are freely
accessible. With the Internet continuing its exponential growth, Arabic Internet content has also been
following the trend, yielding large amounts of textual data available through different Arabic websites. This
paper describes the TALAA corpus, a voluminous general Arabic corpus, built from daily Arabic
newspaper websites. The corpus is a collection of more than 14 million words with 15,891,729 tokens
contained in 57,827 different articles. A part of the TALAA corpus has been tagged to construct an
annotated Arabic corpus of about 7000 tokens, the POS-tagger used containing a set of 58 detailed tags. The
annotated corpus was manually checked by two human experts. The methodology used to construct TALAA
is presented and various metrics are applied to it, showing the usefulness of the corpus. The corpus can be
made available to the scientific community upon authorisation. | nan |
QADI Arabic | [
{
"Name": "AE",
"Dialect": "ar-AE: (Arabic (United Arab Emirates))",
"Volume": "28,011",
"Unit": "sentences"
},
{
"Name": "BH",
"Dialect": "ar-BH: (Arabic (Bahrain))",
"Volume": "28,479",
"Unit": "sentences"
},
{
"Name": "DZ",
"Dialect": "ar-DZ: (Arabic (Algeria))",
"Volume": "17,773",
"Unit": "sentences"
},
{
"Name": "EG",
"Dialect": "ar-EG: (Arabic (Egypt))",
"Volume": "67,983",
"Unit": "sentences"
},
{
"Name": "IQ",
"Dialect": "ar-IQ: (Arabic (Iraq))",
"Volume": "18,545",
"Unit": "sentences"
},
{
"Name": "JO",
"Dialect": "ar-JO: (Arabic (Jordan))",
"Volume": "34,289",
"Unit": "sentences"
},
{
"Name": "KW",
"Dialect": "ar-KW: (Arabic (Kuwait))",
"Volume": "50,153",
"Unit": "sentences"
},
{
"Name": "LB",
"Dialect": "ar-LB: (Arabic (Lebanon))",
"Volume": "38,580",
"Unit": "sentences"
},
{
"Name": "LY",
"Dialect": "ar-LY: (Arabic (Libya))",
"Volume": "41,052",
"Unit": "sentences"
},
{
"Name": "MA",
"Dialect": "ar-MA: (Arabic (Morocco))",
"Volume": "12,991",
"Unit": "sentences"
},
{
"Name": "OM",
"Dialect": "ar-OM: (Arabic (Oman))",
"Volume": "24,955",
"Unit": "sentences"
},
{
"Name": "PL",
"Dialect": "ar-PS: (Arabic (Palestine))",
"Volume": "48,814",
"Unit": "sentences"
},
{
"Name": "QA",
"Dialect": "ar-QA: (Arabic (Qatar))",
"Volume": "36,873",
"Unit": "sentences"
},
{
"Name": "SA",
"Dialect": "ar-SA: (Arabic (Saudi Arabia))",
"Volume": "35,595",
"Unit": "sentences"
},
{
"Name": "SD",
"Dialect": "ar-SD: (Arabic (Sudan))",
"Volume": "16,439",
"Unit": "sentences"
},
{
"Name": "SY",
"Dialect": "ar-SY: (Arabic (Syria))",
"Volume": "18,511",
"Unit": "sentences"
},
{
"Name": "TN",
"Dialect": "ar-TN: (Arabic (Tunisia))",
"Volume": "13,094",
"Unit": "sentences"
},
{
"Name": "YE",
"Dialect": "ar-YE: (Arabic (Yemen))",
"Volume": "11,756",
"Unit": "sentences"
}
] | nan | https://alt.qcri.org/resources/qadi | Apache-2.0 | 2,020 | ar | mixed | social media | text | crawling and annotation(other) | Dialects dataset | 540,590 | sentences | Medium | QCRI | nan | Arabic Dialect Identification in the Wild | https://arxiv.org/pdf/2005.06557.pdf | Arab | No | QCRI Resources | Free | nan | Yes | dialect identification | ArXiv | 16.0 | preprint | ArXiv | Ahmed Abdelali,Hamdy Mubarak,Younes Samih,Sabit Hassan,Kareem Darwish | ,,University Of Düsseldorf;Computational Linguistics,, | We present QADI, an automatically collected dataset of tweets belonging to a wide range of country-level Arabic dialects -covering 18 different countries in the Middle East and North Africa region. Our method for building this dataset relies on applying multiple filters to identify users who belong to different countries based on their account descriptions and to eliminate tweets that are either written in Modern Standard Arabic or contain inappropriate language. The resultant dataset contains 540k tweets from 2,525 users who are evenly distributed across 18 Arab countries. Using intrinsic evaluation, we show that the labels of a set of randomly selected tweets are 91.5% accurate. For extrinsic evaluation, we are able to build effective country-level dialect identification on tweets with a macro-averaged F1-score of 60.6% across 18 classes. | nan |
Arabench | [] | nan | https://alt.qcri.org/resources1/mt/arabench/ | Apache-2.0 | 2,020 | ar | mixed | other | text | other | an evaluation suite for dialectal Arabic to English machine translation | 947,000 | sentences | Low | QCRI | contains data from APT, MDC, MADAR, QCA(QAraC, the bible) | AraBench: Benchmarking Dialectal Arabic-English Machine Translation | https://aclanthology.org/2020.coling-main.447.pdf | Arab | No | QCRI Resources | Free | nan | Yes | machine translation | COLING | 1.0 | conference | International Conference on Computational Linguistics | Hassan Sajjad,Ahmed Abdelali,Nadir Durrani,Fahim Dalvi | ,,Qatar Computing Research Institute, | Low-resource machine translation suffers from the scarcity of training data and the unavailability of standard evaluation sets. While a number of research efforts target the former, the unavailability of evaluation benchmarks remain a major hindrance in tracking the progress in low-resource machine translation. In this paper, we introduce AraBench, an evaluation suite for dialectal Arabic to English machine translation. Compared to Modern Standard Arabic, Arabic dialects are challenging due to their spoken nature, non-standard orthography, and a large variation in dialectness. To this end, we pool together already available Dialectal Arabic-English resources and additionally build novel test sets. AraBench offers 4 coarse, 15 fine-grained and 25 city-level dialect categories, belonging to diverse genres, such as media, chat, religion and travel with varying level of dialectness. We report strong baselines using several training settings: fine-tuning, back-translation and data augmentation. The evaluation suite opens a wide range of research frontiers to push efforts in low-resource machine translation, particularly Arabic dialect translation. The evaluation suite and the dialectal system are publicly available for research purposes. | nan |
Arabic Speech Commands Dataset | [] | https://huggingface.co/datasets/arbml/Speech_Commands_Dataset | https://github.com/abdulkaderghandoura/arabic-speech-commands-dataset | CC BY 4.0 | 2,021 | ar | ar-MSA: (Arabic (Modern Standard Arabic)) | other | spoken | manual curation | This dataset is designed to help train simple machine learning models that serve educational and research purposes in the speech recognition domain | 3 | hours | Low | Multiple institutions | nan | Building and benchmarking an Arabic Speech Commands dataset for small-footprint keyword spotting | https://www.sciencedirect.com/science/article/pii/S0952197621001147 | Arab | No | GitHub | Free | nan | No | speech recognition | EAAI | 0.0 | journal | Engineering Applications of Artificial Intelligence | Abdulkader Ghandoura,Farouk Hjabo,Oumayma Al Dakkak | ,, | The introduction of the Google Speech Commands dataset accelerated research and resulted in a variety of new deep learning approaches that address keyword spotting tasks. The main contribution of this work is the building of an Arabic Speech Commands dataset, a counterpart to Google’s dataset. Our dataset consists of 12000 instances, collected from 30 contributors, and grouped into 40 keywords. We also report different experiments to benchmark this dataset using classical machine learning and deep learning approaches, the best of which is a Convolutional Neural Network with Mel-Frequency Cepstral Coefficients that achieved an accuracy of 98%. Additionally, we point out some key ideas to be considered in such tasks. | nan |
Arabic OSACT4 : Offensive Language Detection | [] | https://huggingface.co/datasets/arbml/OSACT4_hatespeech | https://github.com/motazsaad/arabic-hatespeech-data/blob/master/OSACT4/README.md | unknown | 2,020 | ar | mixed | social media | text | crawling and annotation(other) | OSACT4 Shared Task on Offensive Language Detection | 8,000 | sentences | High | nan | nan | Overview of OSACT4 Arabic Offensive Language Detection Shared Task
| https://aclanthology.org/2020.osact-1.7.pdf | Arab-Latn | No | CodaLab | Free | nan | Yes | offensive language detection | OSACT | 24.0 | workshop | Workshop on Open-Source Arabic Corpora and Processing Tools | Hamdy Mubarak,Kareem Darwish,Walid Magdy,Tamer Elsayed,H. Al-Khalifa | ,,The University of Edinburgh,, | This paper provides an overview of the offensive language detection shared task at the 4th workshop on Open-Source Arabic Corpora and Processing Tools (OSACT4). There were two subtasks, namely: Subtask A, involving the detection of offensive language, which contains unacceptable or vulgar content in addition to any kind of explicit or implicit insults or attacks against individuals or groups; and Subtask B, involving the detection of hate speech, which contains insults or threats targeting a group based on their nationality, ethnicity, race, gender, political or sport affiliation, religious belief, or other common characteristics. In total, 40 teams signed up to participate in Subtask A, and 14 of them submitted test runs. For Subtask B, 33 teams signed up to participate and 13 of them submitted runs. We present and analyze all submissions in this paper. | nan |
Arabic Keyphrase dataset | [] | https://huggingface.co/datasets/arbml/Keyphrase_Extraction | https://github.com/logmani/ArabicDataset | unknown | 2,017 | ar | ar-MSA: (Arabic (Modern Standard Arabic)) | news articles | text | crawling and annotation(other) | A dataset in Arabic language for automatic keyphrase extraction algorithms | 400 | documents | Low | kfupm | nan | ARABIC DATASET FOR AUTOMATIC KEYPHRASE EXTRACTION | https://airccj.org/CSCP/vol7/csit76321.pdf | Arab-Latn | No | GitHub | Free | nan | No | keyphrase extraction | CSIT | 0.0 | conference | International Conference on Computer Science and Information Technologies | Mohammed Al Logmani,H. Muhtaseb | , | We propose a dataset in Arabic language for automatic keyphrase extraction algorithms. Our Arabic dataset contains 400 documents along with their keyphrases. The dataset covers eighteen different categories. An evaluation using a state-of-the-art algorithm demonstrates the accuracy of our dataset is similar to that of English datasets. | nan |
MPOLD: Multi Platforms Offensive Language Dataset | [] | https://huggingface.co/datasets/arbml/MPOLD | https://github.com/shammur/Arabic-Offensive-Multi-Platform-SocialMedia-Comment-Dataset | Apache-2.0 | 2,020 | ar | mixed | social media | text | crawling and annotation(other) | Arabic Offensive Comments dataset from Multiple Social Media Platforms | 400 | documents | Medium | QCRI | nan | A Multi-Platform Arabic News Comment Dataset for Offensive Language Detection | https://aclanthology.org/2020.lrec-1.761.pdf | Arab | No | GitHub | Free | nan | No | offensive language detection | LREC | 10.0 | conference | International Conference on Language Resources and Evaluation | S. A. Chowdhury,Hamdy Mubarak,Ahmed Abdelali,Soon-Gyo Jung,B. Jansen,Joni O. Salminen | ,,,,, | Access to social media often enables users to engage in conversation with limited accountability. This allows a user to share their opinions and ideology, especially regarding public content, occasionally adopting offensive language. This may encourage hate crimes or cause mental harm to targeted individuals or groups. Hence, it is important to detect offensive comments in social media platforms. Typically, most studies focus on offensive commenting in one platform only, even though the problem of offensive language is observed across multiple platforms. Therefore, in this paper, we introduce and make publicly available a new dialectal Arabic news comment dataset, collected from multiple social media platforms, including Twitter, Facebook, and YouTube. We follow two-step crowd-annotator selection criteria for low-representative language annotation task in a crowdsourcing platform. Furthermore, we analyze the distinctive lexical content along with the use of emojis in offensive comments. We train and evaluate the classifiers using the annotated multi-platform dataset along with other publicly available data. Our results highlight the importance of multiple platform dataset for (a) cross-platform, (b) cross-domain, and (c) cross-dialect generalization of classifier performance. | nan |
Arabic RC datasets | [] | https://huggingface.co/datasets/arbml/Arabic_RC_AQA | https://github.com/MariamBiltawi/Arabic_RC_datasets | unknown | 2,020 | ar | ar-MSA: (Arabic (Modern Standard Arabic)) | other | text | crawling and annotation(other) | Arabic Reading Comprehension Benchmarks Created Semiautomatically | 2,862 | sentences | Low | PSUT | AQA (1008), Modified TREC (979), Modified CLEF (418) | Arabic Reading Comprehension Benchmarks Created Semiautomatically | https://www.semanticscholar.org/paper/Arabic-Reading-Comprehension-Benchmarks-Created-Biltawi-Awajan/e637ba939d78e6027bbcc8445f93605d36436421 | Arab | No | GitHub | Free | nan | No | question answering | ACIT | 0.0 | conference | International Arab Conference on Information Technology | Mariam Biltawi,A. Awajan,Sara Tedmori | ,, | Reading comprehension is the task of answering questions from paragraphs; it is also considered a subtask of question-answering systems. Although Arabic language is a language spoken by more than 330 million native speakers, it lacks the required resources, which are needed by the Arabic reading comprehension task to serve as a benchmark dataset. The goal of this work is to present the phases of creating Arabic reading comprehension benchmark dataset semiautomatically. The phases include; data collection, manual check, Google search, document retrieval, and paragraph retrieval. The paper also conducts a thorough evaluation for the created datasets. | nan |
Arabic Satirical Fake News Dataset | [] | https://huggingface.co/datasets/arbml/Satirical_Fake_News | https://github.com/sadanyh/Arabic-Satirical-Fake-News-Dataset | CC BY 4.0 | 2,020 | ar | mixed | other | text | crawling | A Study of Arabic Satirical Fake News | 6,895 | documents | Low | Multiple institutions | ComVE | Fake or Real? A Study of Arabic Satirical Fake News
| https://arxiv.org/pdf/2011.00452.pdf | Arab | No | GitHub | Free | nan | No | fake news detection | RDSM | 4.0 | workshop | International Workshop on Rumours and Deception in Social Media | Hadeel Saadany,Emad Mohamed,Constantin Orasan | ,,University of Surrey, UK | One very common type of fake news is satire which comes in a form of a news website or an online platform that parodies reputable real news agencies to create a sarcastic version of reality. This type of fake news is often disseminated by individuals on their online platforms as it has a much stronger effect in delivering criticism than through a straightforward message. However, when the satirical text is disseminated via social media without mention of its source, it can be mistaken for real news. This study conducts several exploratory analyses to identify the linguistic properties of Arabic fake news with satirical content. It shows that although it parodies real news, Arabic satirical news has distinguishing features on the lexico-grammatical level. We exploit these features to build a number of machine learning models capable of identifying satirical fake news with an accuracy of up to 98.6%. The study introduces a new dataset (3185 articles) scraped from two Arabic satirical news websites (‘Al-Hudood’ and ‘Al-Ahram Al-Mexici’) which consists of fake news. The real news dataset consists of 3710 articles collected from three official news sites: the ‘BBC-Arabic’, the ‘CNN-Arabic’ and ‘Al-Jazeera news’. Both datasets are concerned with political issues related to the Middle East. | nan |
DAWQAS: A Dataset for Arabic Why Question Answering System | [] | https://huggingface.co/datasets/arbml/DAWQAS | https://github.com/masun/DAWQAS | unknown | 2,018 | ar | ar-MSA: (Arabic (Modern Standard Arabic)) | other | text | crawling and annotation(other) | A Dataset for Arabic Why Question Answering System | 3,205 | sentences | Low | Multiple institutions | nan | DAWQAS: A Dataset for Arabic Why Question Answering System | https://www.sciencedirect.com/science/article/pii/S1877050918321690 | Arab | No | GitHub | Free | nan | No | question answering | ACLING | 6.0 | conference | nternational Conference on AI in Computational Linguistics | W. S. Ismail,Masun Nabhan Homsi | , | Abstract A why question answering system is a tool designed to answer why-questions posed in natural language. Several papers have been published on the problem of answering why-questions. In particular, attempts have been made to analyze Arabic text and predict which passages are best candidates for the why-questions; employing different datasets with limited size and not publicly available. To overcome these limitations, this paper introduces the new publicly available dataset, DAWQAS: Dataset for Arabic Why Question Answering System. It consists of 3205 of why question-answer pairs that were first scraped from public Arabic websites, then texts were preprocessed and converted to feature vectors. Afterwards, why-answers were re-categorized based on their domains. Finally, the rhetorical relations’ probabilities based on discourse markers were computed for each sentence in the dataset. DAWQAS is a valuable resource for research and evaluation in language understanding. The new dataset is freely available at https://github.com/masun/DAWQAS . | nan |
The SADID Evaluation Datasets | [
{
"Name": "Levantine",
"Dialect": "ar-LEV: (Arabic(Levant))",
"Volume": "8,988",
"Unit": "sentences"
},
{
"Name": "Egyptian",
"Dialect": "ar-EG: (Arabic (Egypt))",
"Volume": "8,988",
"Unit": "sentences"
},
{
"Name": "MSA",
"Dialect": "ar-MSA: (Arabic (Modern Standard Arabic))",
"Volume": "2,994",
"Unit": "sentences"
},
{
"Name": "English",
"Dialect": "mixed",
"Volume": "8,994",
"Unit": "sentences"
}
] | https://huggingface.co/datasets/arbml/SADID | https://github.com/we7el/SADID | unknown | 2,020 | ar | mixed | other | text | other | Evaluation Datasets for Low-Resource Spoken Language Machine Translation of Arabic Dialects | 29,964 | sentences | Low | Stanford University | Contains curated data and data from the following corpus (LDC2012T09, LDC2019T01, LDC2019T18, LDC2020T05, LDC2012T09) | The SADID Evaluation Datasets for Low-Resource Spoken Language
Machine Translation of Arabic Dialects | https://aclanthology.org/2020.coling-main.530.pdf | Arab | No | GitHub | Free | nan | Yes | machine translation | COLING | 0.0 | conference | International Conference on Computational Linguistics | Wael Abid | nan | Low-resource Machine Translation recently gained a lot of popularity, and for certain languages, it has made great strides. However, it is still difficult to track progress in other languages for which there is no publicly available evaluation data. In this paper, we introduce benchmark datasets for Arabic and its dialects. We describe our design process and motivations and analyze the datasets to understand their resulting properties. Numerous successful attempts use large monolingual corpora to augment low-resource pairs. We try to approach augmentation differently and investigate whether it is possible to improve MT models without any external sources of data. We accomplish this by bootstrapping existing parallel sentences and complement this with multilingual training to achieve strong baselines. | nan |
AQAD: Arabic Question-Answer dataset | [] | https://huggingface.co/datasets/arbml/AQAD | https://github.com/adelmeleka/AQAD | unknown | 2,020 | ar | ar-MSA: (Arabic (Modern Standard Arabic)) | wikipedia | text | crawling | Arabic Questions & Answers dataset | 17,911 | sentences | Low | Alexu | QA from wikipedia (based on SQuAD 2 articles) | AQAD: 17,000+ Arabic Questions for Machine Comprehension of Text | https://www.semanticscholar.org/paper/AQAD%3A-17%2C000%2B-Arabic-Questions-for-Machine-of-Text-Atef-Mattar/d633e0f0a9fdd24c5e3e697478bcc30fc23c8cc8 | Arab | No | GitHub | Free | nan | No | question answering | AICCSA | 0.0 | conference | International Conference on Computer Systems and Applications | Adel Atef,Bassam Mattar,Sandra Sherif,Eman Elrefai,Marwan Torki | ,,,, | Current Arabic Machine Reading for Question Answering datasets suffer from important shortcomings. The available datasets are either small-sized high-quality collections or large-sized low-quality datasets. To address the aforementioned problems we present our Arabic Question-Answer dataset (AQAD). AQAD is a new Arabic reading comprehension large-sized high-quality dataset consisting of 17,000+ questions and answers. To collect the AQAD dataset, we present a fully automated data collector. Our collector works on a set of Arabic Wikipedia articles for the extractive question answering task. The chosen articles match the articles used in the well-known Stanford Question Answering Dataset (SQuAD). We provide evaluation results on the AQAD dataset using two state-of-the-art models for machine-reading question answering problems. Namely, BERT and BIDAF models which result in 0.37 and 0.32 F-1 measure on AQAD dataset. | nan |
ATAR | [] | https://huggingface.co/datasets/arbml/Arabizi_Transliteration | https://github.com/bashartalafha/Arabizi-Transliteration | CC BY-SA | 2,021 | ar | mixed | other | text | manual curation | Arabizi transliteration | 2,743 | tokens | Low | Multiple institutions | nan | Atar: Attention-based LSTM for Arabizi transliteration
| http://ijece.iaescore.com/index.php/IJECE/article/view/22767/14781 | Arab-Latn | No | GitHub | Free | nan | No | transliteration | IJECE | 0.0 | journal | International Journal of Electrical and Computer Engineering | Bashar Talafha,Analle Abuammar,M. Al-Ayyoub | ,, | A non-standard romanization of Arabic script, known as Arbizi, is widely used in Arabic online and SMS/chat communities. However, since state-of-the-art tools and applications for Arabic NLP expects Arabic to be written in Arabic script, handling contents written in Arabizi requires a special attention either by building customized tools or by transliterating them into Arabic script. The latter approach is the more common one and this work presents two significant contributions in this direction. The first one is to collect and publicly release the first large-scale “Arabizi to Arabic script” parallel corpus focusing on the Jordanian dialect and consisting of more than 25 k pairs carefully created and inspected by native speakers to ensure highest quality. Second, we present Atar, an attention-based encoder-decoder model for Arabizi transliteration. Training and testing this model on our dataset yields impressive accuracy (79%) and BLEU score (88.49). | nan |
TUNIZI | [] | nan | https://github.com/chaymafourati/TUNIZI-Sentiment-Analysis-Tunisian-Arabizi-Dataset | unknown | 2,020 | ar | mixed | social media | text | crawling and annotation(other) | first Tunisian Arabizi Dataset including 3K sentences, balanced, covering different topics, preprocessed and annotated as positive and negative | 3,000 | sentences | Medium | iCompass | nan | TUNIZI: A TUNISIAN ARABIZI SENTIMENT ANALYSIS
DATASET | https://arxiv.org/pdf/2004.14303.pdf | Arab-Latn | No | GitHub | Free | nan | No | sentiment analysis | ArXiv | 8.0 | preprint | ArXiv | Chayma Fourati,Abir Messaoudi,Hatem Haddad | ,,iCompass | On social media, Arabic people tend to express themselves in their own local dialects. More particularly, Tunisians use the informal way called "Tunisian Arabizi". Analytical studies seek to explore and recognize online opinions aiming to exploit them for planning and prediction purposes such as measuring the customer satisfaction and establishing sales and marketing strategies. However, analytical studies based on Deep Learning are data hungry. On the other hand, African languages and dialects are considered low resource languages. For instance, to the best of our knowledge, no annotated Tunisian Arabizi dataset exists. In this paper, we introduce TUNIZI a sentiment analysis Tunisian Arabizi Dataset, collected from social networks, preprocessed for analytical studies and annotated manually by Tunisian native speakers. | nan |
TArC | [] | nan | https://github.com/eligugliotta/tarc | unknown | 2,020 | ar | mixed | social media | text | crawling and annotation(other) | flexible and multi-purpose open corpus in order to be a useful support for different types of analyses: computational and linguistics, as well as for NLP tools training | 4,790 | sentences | Low | Stanford University | nan | TArC: Incrementally and Semi-Automatically Collecting a Tunisian
Arabish Corpus
| https://aclanthology.org/2020.lrec-1.770.pdf | Arab-Latn | No | GitHub | Free | nan | No | transliteration | LREC | 2.0 | conference | International Conference on Language Resources and Evaluation | Elisa Gugliotta,Marco Dinarelli | , | This article describes the constitution process of the first morpho-syntactically annotated Tunisian Arabish Corpus (TArC). Arabish, also known as Arabizi, is a spontaneous coding of Arabic dialects in Latin characters and “arithmographs” (numbers used as letters). This code-system was developed by Arabic-speaking users of social media in order to facilitate the writing in the Computer-Mediated Communication (CMC) and text messaging informal frameworks. Arabish differs for each Arabic dialect and each Arabish code-system is under-resourced, in the same way as most of the Arabic dialects. In the last few years, the attention of NLP studies on Arabic dialects has considerably increased. Taking this into consideration, TArC will be a useful support for different types of analyses, computational and linguistic, as well as for NLP tools training. In this article we will describe preliminary work on the TArC semi-automatic construction process and some of the first analyses we developed on TArC. In addition, in order to provide a complete overview of the challenges faced during the building process, we will present the main Tunisian dialect characteristics and its encoding in Tunisian Arabish. | nan |
CheckThat-AR | [] | nan | https://gitlab.com/bigirqu/checkthat-ar/ | unknown | 2,020 | ar | mixed | social media | text | crawling and annotation(other) | check-worthiness datasets | 7,500 | sentences | Medium | nan | nan | Overview of CheckThat! 2020 Arabic: | http://www.dei.unipd.it/~ferro/CLEF-WN-Drafts/CLEF2020/paper_257.pdf | Arab | No | GitLab | Free | nan | Yes | claim verification | CLEF | 9.0 | conference | Conference and Labs of the Evaluation Forum | Maram Hasanain,Fatima Haouari,Reem Suwaileh,Zien Sheikh Ali,Bayan Hamdan,Tamer Elsayed,Alberto Barrón-Cedeño,Giovanni Da San Martino,Preslav Nakov | ,,,,,,,Qatar Computing Research Institute, | In this paper, we make freely accessible ANETAC1 our English-Arabic named entity transliteration and
classification dataset that we built from freely available parallel translation corpora. The dataset contains
79, 924 instances, each instance is a triplet (e, a, c), where e is the English named entity, a is its Arabic
transliteration and c is its class that can be either a Person, a Location, or an Organization. The ANETAC
dataset is mainly aimed for the researchers that are working on Arabic named entity transliteration, but it can
also be used for named entity classification purposes. This dataset was developed and used as part of a previous
research study done by Hadj Ameur et al. | nan |
ANETAC | [] | nan | https://github.com/MohamedHadjAmeur/ANETAC | unknown | 2,020 | ar | ar-MSA: (Arabic (Modern Standard Arabic)) | other | text | crawling and annotation(other) | English-Arabic named entity transliteration and classification dataset | 79,924 | sentences | Low | USTHB University,University of Salford
| nan | Automatic Identification and Verification | https://arxiv.org/pdf/1907.03110.pdf | Arab-Latn | No | GitHub | Free | nan | No | named entity recognition,transliteration | CLEF | 35.0 | conference | Conference and Labs of the Evaluation Forum | Alberto Barrón-Cedeño,Tamer Elsayed,Preslav Nakov,Giovanni Da San Martino,Maram Hasanain,Reem Suwaileh,Fatima Haouari,Nikolay Babulkov,Bayan Hamdan,Alex Nikolov,Shaden Shaar,Zien Sheikh Ali | ,,,Qatar Computing Research Institute,,,,,,,, | We present an overview of the third edition of the CheckThat! Lab at CLEF 2020. The lab featured five tasks in two different languages: English and Arabic. The first four tasks compose the full pipeline of claim verification in social media: Task 1 on check-worthiness estimation, Task 2 on retrieving previously fact-checked claims, Task 3 on evidence retrieval, and Task 4 on claim verification. The lab is completed with Task 5 on check-worthiness estimation in political debates and speeches. A total of 67 teams registered to participate in the lab (up from 47 at CLEF 2019), and 23 of them actually submitted runs (compared to 14 at CLEF 2019). Most teams used deep neural networks based on BERT, LSTMs, or CNNs, and achieved sizable improvements over the baselines on all tasks. Here we describe the tasks setup, the evaluation results, and a summary of the approaches used by the participants, and we discuss some lessons learned. Last but not least, we release to the research community all datasets from the lab as well as the evaluation scripts, which should enable further research in the important tasks of check-worthiness estimation and automatic claim verification. | nan |
AKEC | [] | https://huggingface.co/datasets/arbml/AKEC | https://github.com/ailab-uniud/akec | unknown | 2,016 | ar | ar-MSA: (Arabic (Modern Standard Arabic)) | news articles | text | crawling and annotation(other) | The corpus consists in 160 arabic documents and their keyphrases. | 160 | documents | Low | University of Udine,University of Sheffield | Contains the following corpus Arabic Newspapers Corpus, Corpus of Contemporary Arabic,Essex Arabic Summaries Corpus,Open Source Arabic Corpora | Towards building a standard dataset for Arabic keyphrase extraction evaluation | https://ieeexplore.ieee.org/document/7875927 | Arab | No | GitHub | Free | nan | Yes | keyphrase extraction | IALP | 2.0 | conference | International Conference on Asian Language Processing | Muhammad Helmy,Marco Basaldella,Eddy Maddalena,S. Mizzaro,Gianluca Demartini | ,,,, | Keyphrases are short phrases that best represent a document content. They can be useful in a variety of applications, including document summarization and retrieval models. In this paper, we introduce the first dataset of keyphrases for an Arabic document collection, obtained by means of crowdsourcing. We experimentally evaluate different crowdsourced answer aggregation strategies and validate their performances against expert annotations to evaluate the quality of our dataset. We report about our experimental results, the dataset features, some lessons learned, and ideas for future work. | nan |
AraCust | [] | nan | https://peerj.com/articles/cs-510/#supplemental-information | unknown | 2,021 | ar | ar-SA: (Arabic (Saudi Arabia)) | social media | text | crawling and annotation(other) | Saudi Telecom Tweets corpus for sentiment analysis | 20,000 | sentences | Medium | Durham University,Princess Nourah bint Abdulrahman University | nan | AraCust: a Saudi Telecom Tweets corpus for sentiment analysis | https://peerj.com/articles/cs-510/#supplemental-information | Arab | No | GitHub | Upon-Request | nan | No | sentiment analysis | PeerJ Comput. Sci. | 0.0 | journal | The open access journal for computer science | Latifah Almuqren,A. Cristea | , | Comparing Arabic to other languages, Arabic lacks large corpora for Natural Language Processing (Assiri, Emam & Al-Dossari, 2018; Gamal et al., 2019). A number of scholars depended on translation from one language to another to construct their corpus (Rushdi-Saleh et al., 2011). This paper presents how we have constructed, cleaned, pre-processed, and annotated our 20,0000 Gold Standard Corpus (GSC) AraCust, the first Telecom GSC for Arabic Sentiment Analysis (ASA) for Dialectal Arabic (DA). AraCust contains Saudi dialect tweets, processed from a self-collected Arabic tweets dataset and has been annotated for sentiment analysis, i.e.,manually labelled (k=0.60). In addition, we have illustrated AraCust’s power, by performing an exploratory data analysis, to analyse the features that were sourced from the nature of our corpus, to assist with choosing the right ASA methods for it. To evaluate our Golden Standard corpus AraCust, we have first applied a simple experiment, using a supervised classifier, to offer benchmark outcomes for forthcoming works. In addition, we have applied the same supervised classifier on a publicly available Arabic dataset created from Twitter, ASTD (Nabil, Aly & Atiya, 2015). The result shows that our dataset AraCust outperforms the ASTD result with 91% accuracy and 89% F1avg score. The AraCust corpus will be released, together with code useful for its exploration, via GitHub as a part of this submission. | nan |
Arabic Empathetic Dialogues | [] | nan | https://github.com/aub-mind/Arabic-Empathetic-Chatbot | unknown | 2,020 | ar | ar-MSA: (Arabic (Modern Standard Arabic)) | other | text | manual curation | 38K samples of open-domain utterances and empathetic responses in Modern Standard Arabic (MSA) | 36,629 | sentences | Low | AUB | nan | Empathy-driven Arabic Conversational Chatbot
| https://aclanthology.org/2020.wanlp-1.6.pdf | Arab | No | GitHub | Free | nan | No | dialogue generation | WANLP | 3.0 | workshop | Arabic Natural Language Processing Workshop | Tarek Naous,Christian Hokayem,Hazem M. Hajj | ,, | Conversational models have witnessed a significant research interest in the last few years with the advancements in sequence generation models. A challenging aspect in developing human-like conversational models is enabling the sense of empathy in bots, making them infer emotions from the person they are interacting with. By learning to develop empathy, chatbot models are able to provide human-like, empathetic responses, thus making the human-machine interaction close to human-human interaction. Recent advances in English use complex encoder-decoder language models that require large amounts of empathetic conversational data. However, research has not produced empathetic bots for Arabic. Furthermore, there is a lack of Arabic conversational data labeled with empathy. To address these challenges, we create an Arabic conversational dataset that comprises empathetic responses. However, the dataset is not large enough to develop very complex encoder-decoder models. To address the limitation of data scale, we propose a special encoder-decoder composed of a Long Short-Term Memory (LSTM) Sequence-to-Sequence (Seq2Seq) with Attention. The experiments showed success of our proposed empathy-driven Arabic chatbot in generating empathetic responses with a perplexity of 38.6, an empathy score of 3.7, and a fluency score of 3.92. | nan |
ARCD | [] | https://huggingface.co/datasets/arcd | https://github.com/husseinmozannar/SOQAL | MIT License | 2,019 | ar | ar-MSA: (Arabic (Modern Standard Arabic)) | wikipedia | text | crawling and annotation(other) | 1,395 questions posed by crowdworkers on Wikipedia articles | 1,395 | sentences | Low | AUB | Includes translation of SQuAD version 1.1 | Neural Arabic Question Answering
| https://arxiv.org/pdf/1906.05394.pdf | Arab | No | GitHub | Free | nan | Yes | question answering | WANLP | 29.0 | workshop | Arabic Natural Language Processing Workshop | Hussein Mozannar,Karl El Hajal,Elie Maamary,Hazem M. Hajj | ,,, | This paper tackles the problem of open domain factual Arabic question answering (QA) using Wikipedia as our knowledge source. This constrains the answer of any question to be a span of text in Wikipedia. Open domain QA for Arabic entails three challenges: annotated QA datasets in Arabic, large scale efficient information retrieval and machine reading comprehension. To deal with the lack of Arabic QA datasets we present the Arabic Reading Comprehension Dataset (ARCD) composed of 1,395 questions posed by crowdworkers on Wikipedia articles, and a machine translation of the Stanford Question Answering Dataset (Arabic-SQuAD). Our system for open domain question answering in Arabic (SOQAL) is based on two components: (1) a document retriever using a hierarchical TF-IDF approach and (2) a neural reading comprehension model using the pre-trained bi-directional transformer BERT. Our experiments on ARCD indicate the effectiveness of our approach with our BERT-based reader achieving a 61.3 F1 score, and our open domain system SOQAL achieving a 27.6 F1 score. | nan |
Arabic Text Diacritization | [] | https://huggingface.co/datasets/arbml/arabic_text_diacritization | https://github.com/AliOsm/arabic-text-diacritization | MIT License | 2,019 | ar | ar-MSA: (Arabic (Modern Standard Arabic)) | other | text | other | Arabic Text Diacritization dataset | 55,000 | sentences | Low | JUST | nan | Arabic Text Diacritization Using Deep Neural
Networks | https://arxiv.org/pdf/1905.01965.pdf | Arab | No | GitHub | Free | nan | Yes | diacritization | ICCAIS | 12.0 | conference | International Conference on Computer Applications & Information Security | Ali Fadel,Ibraheem Tuffaha,Bara' Al-Jawarneh,M. Al-Ayyoub | ,,, | Diacritization of Arabic text is both an interesting and a challenging problem at the same time with various applications ranging from speech synthesis to helping students learning the Arabic language. Like many other tasks or problems in Arabic language processing, the weak efforts invested into this problem and the lack of available (open-source) resources hinder the progress towards solving this problem. This work provides a critical review for the currently existing systems, measures and resources for Arabic text diacritization. Moreover, it introduces a much-needed free-for-all cleaned dataset that can be easily used to benchmark any work on Arabic diacritization. Extracted from the Tashkeela Corpus, the dataset consists of 55K lines containing about 2.3M words. After constructing the dataset, existing tools and systems are tested on it. The results of the experiments show that the neural Shakkala system significantly outperforms traditional rule-based approaches and other closed-source tools with a Diacritic Error Rate (DER) of 2.88% compared with 13.78%, which the best DER for the non-neural approach (obtained by the Mishkal tool). | nan |
HAAD | [] | nan | https://github.com/msmadi/HAAD | GPL-2.0 | 2,015 | ar | ar-MSA: (Arabic (Modern Standard Arabic)) | books | text | crawling and annotation(other) | Human Annotated Arabic Dataset of Book Reviews for Aspect Based Sentiment Analysis | 2,389 | sentences | Low | JUST | nan | Human Annotated Arabic Dataset of Book Reviews for Aspect Based Sentiment Analysis | https://ieeexplore.ieee.org/document/7300895 | Arab | No | GitHub | Upon-Request | nan | Yes | sentiment analysis | FiCloud | 68.0 | conference | Conference on Future Internet of Things and Cloud | Mohammad Al-Smadi,Omar Qawasmeh,Bashar Talafha,Muhannad Quwaider | ,,, | With the prominent advances in Web interaction and the enormous growth in user-generated content, sentiment analysis has gained more interest in commercial and academic purposes. Recently, sentiment analysis of Arabic user-generated content is increasingly viewed as an important research field. However, the majority of available approaches target the overall polarity of the text. To the best of our knowledge, there is no available research on aspect-based sentiment analysis (ABSA) of Arabic text. This can be explained due to the lack of publically available datasets prepared for ABSA, and to the slow progress in sentiment analysis of Arabic text research in general. This paper fosters the domain of Arabic ABSA, and provides a benchmark human annotated Arabic dataset (HAAD). HAAD consists of books reviews in Arabic which have been annotated by humans with aspect terms and their polarities. Nevertheless, the paper reports a baseline results and a common evaluation technique to facilitate future evaluation of research and methods. | nan |
COVID-19-Arabic-Tweets-Dataset | [] | https://huggingface.co/datasets/arbml/COVID_19_Arabic_Tweets_Dataset | https://github.com/SarahAlqurashi/COVID-19-Arabic-Tweets-Dataset | CC BY-NC-SA 4.0 | 2,020 | ar | mixed | social media | text | crawling | collection of Arabic tweets IDs related to novel coronavirus COVID-19. | 3,934,610 | sentences | Medium | Umm Al-Qura University | nan | Large Arabic Twitter Dataset on COVID-19
| https://arxiv.org/pdf/2004.04315.pdf | Arab-Latn | No | GitHub | Free | nan | Yes | behaviour analysis | ArXiv | 26.0 | preprint | ArXiv | S. Alqurashi,Ahmad Alhindi,E. Alanazi | ,, | The 2019 coronavirus disease (COVID-19), emerged late December 2019 in China, is now rapidly spreading across the globe. At the time of writing this paper, the number of global confirmed cases has passed two millions and half with over 180,000 fatalities. Many countries have enforced strict social distancing policies to contain the spread of the virus. This have changed the daily life of tens of millions of people, and urged people to turn their discussions online, e.g., via online social media sites like Twitter. In this work, we describe the first Arabic tweets dataset on COVID-19 that we have been collecting since January 1st, 2020. The dataset would help researchers and policy makers in studying different societal issues related to the pandemic. Many other tasks related to behavioral change, information sharing, misinformation and rumors spreading can also be analyzed. | nan |
Aljazeera Deleted Comments | [] | nan | https://alt.qcri.org/people/hmubarak/public_html/offensive/ | unknown | 2,017 | ar | mixed | commentary | text | other | offensive and obsene language dataset | 33,100 | sentences | Low | QCRI | nan | Abusive Language Detection on Arabic Social Media
| https://aclanthology.org/W17-3008.pdf | Arab | No | QCRI Resources | Free | nan | Yes | hate speech detection, abusive language detection | ALW | 148.0 | workshop | Abusive Language Online | Hamdy Mubarak,Kareem Darwish,Walid Magdy | ,,The University of Edinburgh | In this paper, we present our work on detecting abusive language on Arabic social media. We extract a list of obscene words and hashtags using common patterns used in offensive and rude communications. We also classify Twitter users according to whether they use any of these words or not in their tweets. We expand the list of obscene words using this classification, and we report results on a newly created dataset of classified Arabic tweets (obscene, offensive, and clean). We make this dataset freely available for research, in addition to the list of obscene words and hashtags. We are also publicly releasing a large corpus of classified user comments that were deleted from a popular Arabic news site due to violations the site’s rules and guidelines. | nan |
Anti-Social Behaviour in Online Communication | [] | nan | https://onedrive.live.com/?authkey=!ACDXj_ZNcZPqzy0&id=6EF6951FBF8217F9!105&cid=6EF6951FBF8217F9 | unknown | 2,018 | ar | mixed | social media | text | crawling and annotation(other) | a corpus of 15,050 labelled YouTube comments in Arabic | 15,050 | sentences | Medium | Limrick Uni | nan | Detection of Anti-Social Behaviour
in Online Communication in
Arabic | https://ulir.ul.ie/bitstream/handle/10344/9946/Alakrot_2019_Detection.pdf?sequence=2 | Arab-Latn | No | OneDrive | Free | nan | No | behaviour analysis | ACLING | 33.0 | conference | nternational Conference on AI in Computational Linguistics | Azalden Alakrot,Liam Murray,Nikola S. Nikolov | ,, | Abstract Warning: this paper contains a range of words which may cause offence. In recent years, many studies target anti-social behaviour such as offensive language and cyberbullying in online communication. Typically, these studies collect data from various reachable sources, the majority of the datasets being in English. However, to the best of our knowledge, there is no dataset collected from the YouTube platform targeting Arabic text and overall there are only a few datasets of Arabic text, collected from other social platforms for the purpose of offensive language detection. Therefore, in this paper we contribute to this field by presenting a dataset of YouTube comments in Arabic, specifically designed to be used for the detection of offensive language in a machine learning scenario. Our dataset contains a range of offensive language and flaming in the form of YouTube comments. We document the labelling process we have conducted, taking into account the difference in the Arab dialects and the diversity of perception of offensive language throughout the Arab world. Furthermore, statistical analysis of the dataset is presented, in order to make it ready for use as a training dataset for predictive modeling. | nan |
MLMA hate speech | [] | https://huggingface.co/datasets/arbml/MLMA_hate_speech_ar | https://github.com/HKUST-KnowComp/MLMA_hate_speech | MIT License | 2,019 | ar | mixed | social media | text | crawling and annotation(other) | Multilingual and Multi-Aspect Hate Speech Analysis | 3,354 | sentences | High | HKUST | nan | Multilingual and Multi-Aspect Hate Speech Analysis
| https://aclanthology.org/D19-1474.pdf | Arab | No | GitHub | Free | nan | No | hate speech detection, abusive language detection | EMNLP | 57.0 | conference | Conference on Empirical Methods in Natural Language Processing | N. Ousidhoum,Zizheng Lin,Hongming Zhang,Y. Song,D. Yeung | ,,,, | Current research on hate speech analysis is typically oriented towards monolingual and single classification tasks. In this paper, we present a new multilingual multi-aspect hate speech analysis dataset and use it to test the current state-of-the-art multilingual multitask learning approaches. We evaluate our dataset in various classification settings, then we discuss how to leverage our annotations in order to improve hate speech detection and classification in general. | nan |
L-HSAB | [] | https://huggingface.co/datasets/arbml/L_HSAB | https://github.com/Hala-Mulki/L-HSAB-First-Arabic-Levantine-HateSpeech-Dataset | unknown | 2,019 | ar | ar-LEV: (Arabic(Levant)) | social media | text | crawling and annotation(other) | Arabic Levantine Hate Speech and Abusive Language Dataset | 5,851 | sentences | High | Multiple institutions | nan | L-HSAB: A Levantine Twitter Dataset for Hate Speech and Abusive Language | https://aclanthology.org/W19-3512.pdf | Arab | No | GitHub | Free | nan | Yes | hate speech detection, abusive language detection | ALW | 46.0 | workshop | Abusive Language Online | Hala Mulki,Hatem Haddad,Chedi Bechikh Ali,Halima Alshabani | ,iCompass,, | ∗Department of Computer Engineering, Konya Technical University, Turkey †RIADI Laboratory, National School of Computer Sciences, University of Manouba, Tunisia ∗∗LISI Laboratory, INSAT, Carthage University, Tunisia ∗∗∗Department of Computer Engineering, Kırıkkale University, Turkey §iCompass Consulting, Tunisia halamulki@selcuk.edu.tr,haddad.Hatem@gmail.com chedi.bechikh@gmail.com,halima.alshabani@gmail.com Abstract | nan |
Large Multi-Domain Resources for Arabic Sentiment Analysis | [] | https://huggingface.co/datasets/arbml/ATT | https://github.com/hadyelsahar/large-arabic-sentiment-analysis-resouces | unknown | 2,015 | ar | mixed | reviews | text | crawling and annotation(other) | Large Multi-Domain Resources for Arabic Sentiment Analysis | 45,498 | sentences | Low | Nile University | nan | Building Large Arabic Multi-domain Resources for Sentiment Analysis | https://link.springer.com/chapter/10.1007/978-3-319-18117-2_2 | Arab | No | GitHub | Free | nan | No | sentiment analysis | CICLing | 127.0 | conference | International Conference on Computational Linguistics and Intelligent Text Processing | Hady ElSahar,S. El-Beltagy | , | While there has been a recent progress in the area of Arabic Sentiment Analysis, most of the resources in this area are either of limited size, domain specific or not publicly available. In this paper, we address this problem by generating large multi-domain datasets for Sentiment Analysis in Arabic. The datasets were scrapped from different reviewing websites and consist of a total of 33K annotated reviews for movies, hotels, restaurants and products. Moreover we build multi-domain lexicons from the generated datasets. Different experiments have been carried out to validate the usefulness of the datasets and the generated lexicons for the task of sentiment classification. From the experimental results, we highlight some useful insights addressing: the best performing classifiers and feature representation methods, the effect of introducing lexicon based features and factors affecting the accuracy of sentiment classification in general. All the datasets, experiments code and results have been made publicly available for scientific purposes. | nan |
TEAD | [] | nan | https://github.com/HSMAabdellaoui/TEAD | GPL-3.0 | 2,018 | ar | mixed | social media | text | crawling and annotation(other) | dataset for Arabic Sentiment Analysis | 6,000,000 | sentences | Medium | Multiple institutions | nan | Using Tweets and Emojis to Build TEAD: an Arabic Dataset for Sentiment Analysis | https://www.researchgate.net/publication/328105014_Using_Tweets_and_Emojis_to_Build_TEAD_an_Arabic_Dataset_for_Sentiment_Analysis | Arab | No | GitHub | Upon-Request | nan | No | sentiment analysis | Computación y Sistemas | 17.0 | journal | Computación y Sistemas | Houssem Abdellaoui,M. Zrigui | , | Our paper presents a distant supervision algorithm for automatically collecting and labeling ‘TEAD’, a dataset for Arabic Sentiment Analysis (SA), using emojis and sentiment lexicons. The data was gathered from Twitter during the period between the 1st of June and the 30th of November 2017. Although the idea of using emojis to collect and label training data for SA, is not novel, getting this approach to work for Arabic dialect was very challenging. We ended up with more than 6 million tweets labeled as Positive, Negative or Neutral. We present the algorithm used to deal with mixed-content tweets (Modern Standard Arabic MSA and Dialect Arabic DA). We also provide properties and statistics of the dataset along side experiments results. Our try outs covered a wide range of standard classifiers proved to be efficient for sentiment classification problem. | nan |
ASTD | [] | https://huggingface.co/datasets/arbml/ASTD | https://github.com/mahmoudnabil/ASTD | GPL-2.0 | 2,015 | ar | mixed | social media | text | crawling and annotation(other) | 10k Arabic sentiment tweets classified into four classes subjective positive, subjective negative, subjective mixed, and objective | 10,006 | sentences | Medium | Cairo University | nan | ASTD: Arabic Sentiment Tweets Dataset
| https://aclanthology.org/D15-1299.pdf | Arab | No | GitHub | Free | nan | Yes | sentiment analysis | EMNLP | 178.0 | conference | Conference on Empirical Methods in Natural Language Processing | Mahmoud Nabil,Mohamed A. Aly,A. Atiya | ,, | This paper introduces ASTD, an Arabic social sentiment analysis dataset gathered from Twitter. It consists of about 10,000 tweets which are classified as objective, subjective positive, subjective negative, and subjective mixed. We present the properties and the statistics of the dataset, and run experiments using standard partitioning of the dataset. Our experiments provide benchmark results for 4 way sentiment classification on the dataset. | nan |
DART | [
{
"Name": "EGY",
"Dialect": "ar-EG: (Arabic (Egypt))",
"Volume": "5,265",
"Unit": "sentences"
},
{
"Name": "GLF",
"Dialect": "ar-GLF: (Arabic (Gulf))",
"Volume": "5,893",
"Unit": "sentences"
},
{
"Name": "IRQ",
"Dialect": "ar-IQ: (Arabic (Iraq))",
"Volume": "5,253",
"Unit": "sentences"
},
{
"Name": "LEV",
"Dialect": "ar-LEV: (Arabic(Levant))",
"Volume": "3,939",
"Unit": "sentences"
},
{
"Name": "MGH",
"Dialect": "ar-NOR: (Arabic (North Africa))",
"Volume": "3,930",
"Unit": "sentences"
}
] | nan | https://www.dropbox.com/s/jslg6fzxeu47flu/DART.zip?dl=0 | unknown | 2,018 | ar | mixed | social media | text | crawling and annotation(other) | Dialectal Arabic Tweets | 24,280 | sentences | Medium | Qatar University | nan | DART: A Large Dataset of Dialectal Arabic Tweets
| https://aclanthology.org/L18-1579.pdf | Arab | No | Dropbox | Free | nan | Yes | dialect identification | LREC | 15.0 | conference | International Conference on Language Resources and Evaluation | Israa Alsarsour,Esraa Mohamed,Reem Suwaileh,T. Elsayed | ,,, | In this paper, we present a new large manually-annotated multi-dialect dataset of Arabic tweets that is publicly available. The Dialectal ARabic Tweets (DART) dataset has about 25K tweets that are annotated via crowdsourcing and it is well-balanced over five main groups of Arabic dialects: Egyptian, Maghrebi, Levantine, Gulf, and Iraqi. The paper outlines the pipeline of constructing the dataset from crawling tweets that match a list of dialect phrases to annotating the tweets by the crowd. We also touch some challenges that we face during the process. We evaluate the quality of the dataset from two perspectives: the inter-annotator agreement and the accuracy of the final labels. Results show that both measures were substantially high for the Egyptian, Gulf, and Levantine dialect groups, but lower for the Iraqi andMaghrebi dialects, which indicates the difficulty of identifying those two dialectsmanually and hence automatically. | nan |
PADIC: Parallel Arabic DIalect Corpus | [
{
"Name": "MSA",
"Dialect": "ar-MSA: (Arabic (Modern Standard Arabic))",
"Volume": "8,244",
"Unit": "sentences"
},
{
"Name": "ALG",
"Dialect": "ar-DZ: (Arabic (Algeria))",
"Volume": "8,244",
"Unit": "sentences"
},
{
"Name": "ANB",
"Dialect": "ar-DZ: (Arabic (Algeria))",
"Volume": "8,244",
"Unit": "sentences"
},
{
"Name": "TUN",
"Dialect": "ar-TN: (Arabic (Tunisia))",
"Volume": "8,244",
"Unit": "sentences"
},
{
"Name": "PAL",
"Dialect": "ar-PS: (Arabic (Palestine))",
"Volume": "8,244",
"Unit": "sentences"
},
{
"Name": "SYR",
"Dialect": "ar-SY: (Arabic (Syria))",
"Volume": "8,244",
"Unit": "sentences"
}
] | https://huggingface.co/datasets/arbml/PADIC | https://smart.loria.fr/corpora/ | unknown | 2,014 | ar | mixed | social media | text | crawling and annotation(other) | s composed of about 6400 sentences of dialects from both the Maghreb and the middle east | 12,824 | sentences | Medium | Loris Fr | nan | A multidialectal parallel corpus of Arabic | https://hal.archives-ouvertes.fr/hal-01261587/document | Arab | No | other | Free | nan | No | machine translation | LREC | 82.0 | conference | International Conference on Language Resources and Evaluation | Houda Bouamor,Nizar Habash,Kemal Oflazer | ,, | The daily spoken variety of Arabic is often termed the colloquial or dialect form of Arabic. There are many Arabic dialects across the Arab World and within other Arabic speaking communities. These dialects vary widely from region to region and to a lesser extent from city to city in each region. The dialects are not standardized, they are not taught, and they do not have official status. However they are the primary vehicles of communication (face-to-face and recently, online) and have a large presence in the arts as well. In this paper, we present the first multidialectal Arabic parallel corpus, a collection of 2,000 sentences in Standard Arabic, Egyptian, Tunisian, Jordanian, Palestinian and Syrian Arabic, in addition to English. Such parallel data does not exist naturally, which makes this corpus a very valuable resource that has many potential applications such as Arabic dialect identification and machine translation. | nan |
MetRec | [] | https://huggingface.co/datasets/metrecc | https://github.com/zaidalyafeai/MetRec | MIT License | 2,020 | ar | ar-CLS: (Arabic (Classic)) | other | text | crawling | More than 40K of verses with their meters | 47,124 | sentences | Low | kfupm | nan | MetRec: A dataset for meter classification of arabic poetry | https://www.sciencedirect.com/science/article/pii/S2352340920313792 | Arab | No | GitHub | Free | nan | Yes | meter classification | Data in brief | 1.0 | journal | Data in brief | Maged S. Al-shaibani,Zaid Alyafeai,Irfan Ahmad | ,, | In this data article, we report a dataset related to the research titled “Meter Classification of Arabic Poems Using Deep Bidirectional Recurrent Neural Networks”[2]. The dataset was collected from a large repository of Arabic poems, Aldiwan website [1]. The data collection was done using a Python script that scrapes the website to find the poems and their associated meters. The dataset contains the verses and their corresponding meter classes. Meter classes are represented as numbers from 0 to 13. The dataset can be highly useful for further research in order to improve the field of Arabic poems’ meter classification. | nan |
AraSenti | [] | nan | https://github.com/nora-twairesh/AraSenti | GPL-3.0 | 2,017 | ar | ar-SA: (Arabic (Saudi Arabia)) | social media | text | crawling and annotation(other) | The corpus contains 17,573 tweets labelled with four labels for sentiment: positive, negative, neutral and mixed | 17,573 | sentences | Medium | KSU | nan | AraSenTi-Tweet: A Corpus for Arabic Sentiment Analysis of Saudi Tweets | https://www.sciencedirect.com/science/article/pii/S1877050917321518 | Arab-Latn | Yes | GitHub | Free | nan | No | sentiment analysis | ACLING | 56.0 | conference | nternational Conference on AI in Computational Linguistics | Nora Al-Twairesh,Hend Suliman Al-Khalifa,A. Al-Salman,Y. Al-Ohali | ,,, | Abstract Arabic Sentiment Analysis is an active research area these days. However, the Arabic language still lacks sufficient language resources to enable the tasks of sentiment analysis. In this paper, we present the details of collecting and constructing a large dataset of Arabic tweets. The techniques used in cleaning and pre-processing the collected dataset are explained. A corpus of Arabic tweets annotated for sentiment analysis was extracted from this dataset. The corpus consists mainly of tweets written in Modern Standard Arabic and the Saudi dialect. The corpus was manually annotated for sentiment. The annotation process is explained in detail and the challenges during the annotation are highlighted. The corpus contains 17,573 tweets labelled with four labels for sentiment: positive, negative, neutral and mixed. Baseline experiments were conducted to provide benchmark results for future work. | Maraim Masoud |
OCA: Opinion corpus for Arabic | [] | nan | http://150.214.174.171:8059/investigacion/recursos/oca-corpus | unknown | 2,011 | ar | mixed | books | text | crawling and annotation(other) | The corpus contains 500 movie reviews collected from different web pages and blogs in Arabic, 250 of them considered as positive reviews, and the other 250 as negative opinions | 500 | sentences | Low | nan | nan | OCA: Opinion corpus for Arabic
| https://onlinelibrary.wiley.com/doi/full/10.1002/asi.21598 | Arab | No | other | Upon-Request | nan | No | sentiment analysis | JASIST | 242.0 | journal | The Journal of the Association for Information Science and Technology | Mohammed Rushdi-Saleh,M. T. Martín-Valdivia,L. A. U. López,José Manuel Perea Ortega | ,,, | Sentiment analysis is a challenging new task related to text mining and natural language processing. Although there are, at present, several studies related to this theme, most of these focus mainly on English texts. The resources available for opinion mining (OM) in other languages are still limited. In this article, we present a new Arabic corpus for the OM task that has been made available to the scientific community for research purposes. The corpus contains 500 movie reviews collected from different web pages and blogs in Arabic, 250 of them considered as positive reviews, and the other 250 as negative opinions. Furthermore, different experiments have been carried out on this corpus, using machine learning algorithms such as support vector machines and Nave Bayes. The results obtained are very promising and we are encouraged to continue this line of research. © 2011 Wiley Periodicals, Inc. | Maraim Masoud |
Arabic Common Voice | [] | https://huggingface.co/datasets/common_voice | https://commonvoice.mozilla.org/ar/datasets | CC0
| 2,020 | ar | ar-MSA: (Arabic (Modern Standard Arabic)) | transcribed audio | spoken | other | an open source, multi-language dataset of voices that anyone can use to train speech-enabled applications. | 85 | hours | Low | mozilla | nan | Common Voice: A Massively-Multilingual Speech Corpus
| https://arxiv.org/pdf/1912.06670.pdf | Arab | No | Mozilla | Upon-Request | nan | Yes | speech recognition | LREC | 131.0 | conference | International Conference on Language Resources and Evaluation | Rosana Ardila,Megan Branson,Kelly Davis,Michael Henretty,Michael Kohler,Josh Meyer,Reuben Morais,Lindsay Saunders,Francis M. Tyers,Gregor Weber | ,,,,,,,,, | The Common Voice corpus is a massively-multilingual collection of transcribed speech intended for speech technology research and development. Common Voice is designed for Automatic Speech Recognition purposes but can be useful in other domains (e.g. language identification). To achieve scale and sustainability, the Common Voice project employs crowdsourcing for both data collection and data validation. The most recent release includes 29 languages, and as of November 2019 there are a total of 38 languages collecting data. Over 50,000 individuals have participated so far, resulting in 2,500 hours of collected audio. To our knowledge this is the largest audio corpus in the public domain for speech recognition, both in terms of number of hours and number of languages. As an example use case for Common Voice, we present speech recognition experiments using Mozilla’s DeepSpeech Speech-to-Text toolkit. By applying transfer learning from a source English model, we find an average Character Error Rate improvement of 5.99 ± 5.48 for twelve target languages (German, French, Italian, Turkish, Catalan, Slovenian, Welsh, Irish, Breton, Tatar, Chuvash, and Kabyle). For most of these languages, these are the first ever published results on end-to-end Automatic Speech Recognition. | Zaid Alyafeai |
MLQA | [] | nan | https://github.com/facebookresearch/mlqa | CC BY-SA 3.0 | 2,020 | multilingual | ar-MSA: (Arabic (Modern Standard Arabic)) | other | text | crawling and annotation(other) | 5K extractive QA instances (12K in English) in SQuAD format in seven languages - English, Arabic, German, Spanish, Hindi, Vietnamese and Simplified Chinese. | 5,852 | documents | Low | Facebook | nan | MLQA: Evaluating Cross-lingual Extractive Question Answering
| https://arxiv.org/pdf/1910.07475.pdff | Arab | No | GitHub | Free | nan | Yes | question answering | ACL | 124.0 | conference | Assofications of computation linguisitcs | Patrick Lewis,Barlas Oğuz,Ruty Rinott,S. Riedel,Holger Schwenk | Facebook AI Research;University College London,Facebook AI,,, | Question answering (QA) models have shown rapid progress enabled by the availability of large, high-quality benchmark datasets. Such annotated datasets are difficult and costly to collect, and rarely exist in languages other than English, making building QA systems that work well in other languages challenging. In order to develop such systems, it is crucial to invest in high quality multilingual evaluation benchmarks to measure progress. We present MLQA, a multi-way aligned extractive QA evaluation benchmark intended to spur research in this area. MLQA contains QA instances in 7 languages, English, Arabic, German, Spanish, Hindi, Vietnamese and Simplified Chinese. MLQA has over 12K instances in English and 5K in each other language, with each instance parallel between 4 languages on average. We evaluate state-of-the-art cross-lingual models and machine-translation-based baselines on MLQA. In all cases, transfer results are shown to be significantly behind training-language performance. | Zaid Alyafeai |
TYDIQA | [] | https://huggingface.co/datasets/tydiqa | https://github.com/google-research-datasets/tydiqa | Apache-2.0 | 2,020 | multilingual | ar-MSA: (Arabic (Modern Standard Arabic)) | other | text | crawling and annotation(other) | question answering dataset covering 11 typologically diverse languages with 200K question-answer pairs | 25,893 | sentences | Low | Google | nan | TYDI QA: A Benchmark for Information-Seeking Question Answering in Typologically Diverse Languages | https://storage.googleapis.com/tydiqa/tydiqa.pdf | Arab | No | GitHub | Free | nan | Yes | question answering | TACL | 91.0 | journal | Transactions of the Association for Computational Linguistics | J. Clark,Eunsol Choi,Michael Collins,Dan Garrette,T. Kwiatkowski,Vitaly Nikolaev,Jennimaria Palomaki | ,,,Google Research,,, | Abstract Confidently making progress on multilingual modeling requires challenging, trustworthy evaluations. We present TyDi QA—a question answering dataset covering 11 typologically diverse languages with 204K question-answer pairs. The languages of TyDi QA are diverse with regard to their typology—the set of linguistic features each language expresses—such that we expect models performing well on this set to generalize across a large number of the world’s languages. We present a quantitative analysis of the data quality and example-level qualitative linguistic analyses of observed language phenomena that would not be found in English-only corpora. To provide a realistic information-seeking task and avoid priming effects, questions are written by people who want to know the answer, but don’t know the answer yet, and the data is collected directly in each language without the use of translation. | Zaid Alyafeai |
BRAD 2.0 | [] | nan | https://github.com/elnagara/BRAD-Arabic-Dataset | unknown | 2,018 | ar | mixed | reviews | text | crawling | extension to BRAD 1.0 with more than 200K extra records to account for several Arabic dialects. | 692,586 | sentences | Low | University of Sharjah | nan | An Annotated Huge Dataset for Standard and Colloquial Arabic Reviews for Subjective Sentiment Analysis | https://www.sciencedirect.com/science/article/pii/S1877050918321781 | Arab | No | GitHub | Free | nan | Yes | review classification | ACLING | 28.0 | conference | nternational Conference on AI in Computational Linguistics | Ashraf Elnagar,Leena Lulu,Omar Einea | ,, | Abstract Sentiment analysis is getting increasingly popular as it facilitates gaining an indication of the wider public opinions or attitudes towards certain products, services, articles, etc. Many researchers have shown considerable interest in this field. Most of these studies have focused on English and other Indo-European languages. Very few studies have addressed the problem for the Arabic language. This is, mostly, due to the rare or nonexistent huge and free Arabic datasets that contains both Modern Standard Arabic (MSA) as well as Dialectal Arabic (DA). Generally, one of the main challenges for developing robust sentiment analysis systems is the availability of such large-scale datasets. Such datasets exist in abundance for English language, while it is not the case for a low-resource language such as the Arabic language. Recently, there have been some efforts for providing relatively large-scale Arabic datasets dedicated for sentiment analysis such as LABR and most recently BRAD 1.0, which is considered as the largest Arabic Book Reviews dataset for sentiment analysis and machine learning applications. In this work, we present BRAD 2.0, an extension to BRAD 1.0 with more than 200K extra records to account for several Arabic dialects. BRAD 2.0 has a total number of 692586 annotated reviews; each represents a single review along with the reviewer’s rating ranging from 1 to 5 of a certain book. The most interesting property of BRAD 2.0 is that it combines both MSA and DA. To verify and validate the proposed dataset, we implement several state-of-the-art supervised and unsupervised classifiers to categorize book reviews. For the unsupervised classifiers, we implemented several models of CNN and RNN classifiers utilizing GloVe-based word embeddings. Although all classifiers performed well, the highest accuracies attained are between 90% and 91%. Experimental results show that BRAD 2.0 is rich and robust. Our key contribution is to make this benchmark-dataset available and accessible to promote further research in the field of Arabic computational linguistic. | Zaid Alyafeai |
Disease NER | [] | https://huggingface.co/datasets/arbml/Disease_NER | https://www.mdpi.com/2306-5729/5/3/60/htm#app1-data-05-00060 | unknown | 2,020 | ar | ar-MSA: (Arabic (Modern Standard Arabic)) | other | text | other | The data consist of 27 Arabic medical articles, totaling around 50,000 words. | 62,506 | tokens | Low | Jouf University | BRAD 1.0 | An Arabic Dataset for Disease Named Entity Recognition with Multi-Annotation Schemes | https://www.mdpi.com/2306-5729/5/3/60/htm | Arab | No | MDPI | Free | nan | No | named entity recognition | Data | 2.0 | journal | nan | Nasser O. Alshammari,S. Alanazi | , | This article outlines a novel data descriptor that provides the Arabic natural language processing community with a dataset dedicated to named entity recognition tasks for diseases. The dataset comprises more than 60 thousand words, which were annotated manually by two independent annotators using the inside–outside (IO) annotation scheme. To ensure the reliability of the annotation process, the inter-annotator agreements rate was calculated, and it scored 95.14%. Due to the lack of research efforts in the literature dedicated to studying Arabic multi-annotation schemes, a distinguishing and a novel aspect of this dataset is the inclusion of six more annotation schemes that will bridge the gap by allowing researchers to explore and compare the effects of these schemes on the performance of the Arabic named entity recognizers. These annotation schemes are IOE, IOB, BIES, IOBES, IE, and BI. Additionally, five linguistic features, including part-of-speech tags, stopwords, gazetteers, lexical markers, and the presence of the definite article, are provided for each record in the dataset. | Zaid Alyafeai |
SNAD | [] | https://huggingface.co/datasets/arbml/SNAD | https://drive.google.com/file/d/1uwD56jaVIbsQQWVqqyL08TgjuTraYFJC/view | unknown | 2,020 | ar | ar-MSA: (Arabic (Modern Standard Arabic)) | news articles | text | crawling | The data was scraped and collected from two of the most famous news sources in Saudi Arabia | 45,000 | documents | Low | Prince Sultan University | nan | SNAD Arabic Dataset for Deep Learning
| https://link.springer.com/chapter/10.1007/978-3-030-55180-3_47 | Arab | No | Gdrive | Free | nan | No | topic classification | IntelliSys | 1.0 | conference | Intelligent Systems Conference | Deem Alsaleh,Mashael Bin Alamir,Souad Larabi Marie-Sainte | ,, | Natural language processing (NLP) captured the attention of researchers for the last years. NLP is applied in various applications and several disciplines. Arabic is a language that also benefited from NLP. However, only few Arabic datasets are available for researchers. For that, applying the Arabic NLP is limited in these datasets. Hence, this paper introduces a new dataset, SNAD. SNAD is collected to fill the gap in Arabic datasets, especially for classification using deep learning. The dataset has more than 45,000 records. Each record consists of the news title, news details, in addition to the news class. The dataset has six different classes. Moreover, cleaning and preprocessing are applied to the raw data to make it more efficient for classification purpose. Finally, the dataset is validated using the Convolutional Neural Networks and the result is efficient. The dataset is freely available online. | Zaid Alyafeai |
NADA | [] | nan | https://www.researchgate.net/publication/326060650_NADA_A_New_Arabic_Dataset | unknown | 2,018 | ar | ar-MSA: (Arabic (Modern Standard Arabic)) | news articles | text | crawling | NADA corpus is collected from two existing corpora,
which are Diab Dataset DAA corpus and OSAC corpus. | 13,066 | documents | Low | Multiple institutions | nan | NADA: New Arabic Dataset for Text Classification
| https://thesai.org/Downloads/Volume9No9/Paper_28-NADA_New_Arabic_Dataset_for_Text_Classification.pdf | Arab | No | ResearchGate | Free | nan | Yes | topic classification | IJACSA | 8.0 | journal | International Journal of Advanced Computer Science and Applications | Nada Alalyani,Souad Larabi Marie-Sainte | , | In the recent years, Arabic Natural Language Processing, including Text summarization, Text simplification, Text Categorization and other Natural Language-related disciplines, are attracting more researchers. Appropriate resources for Arabic Text Categorization are becoming a big necessity for the development of this research. The few existing corpora are not ready for use, they require preprocessing and filtering operations. In addition, most of them are not organized based on standard classification methods which makes unbalanced classes and thus reduced the classification accuracy. This paper proposes a New Arabic Dataset (NADA) for Text Categorization purpose. This corpus is composed of two existing corpora OSAC and DAA. The new corpus is preprocessed and filtered using the recent state of the art methods. It is also organized based on Dewey decimal classification scheme and Synthetic Minority Over-Sampling Technique. The experiment results show that NADA is an efficient dataset ready for use in Arabic Text Categorization. | Zaid Alyafeai |
Arabic Sentiment Lexicons | [] | https://huggingface.co/datasets/arbml/Sentiment_Lexicons | https://saifmohammad.com/WebPages/ArabicSA.html | unknown | 2,016 | ar | mixed | social media | text | crawling and annotation(translation) | by using distant supervision techniques on Arabic tweets, and by translating English sentiment lexicons into Arabic using a freely available statistical machine translation system | 176,364 | tokens | Medium | Multiple institutions | nan | Sentiment Lexicons for Arabic Social Media
| https://aclanthology.org/L16-1006.pdf | Arab | No | other | Free | nan | No | sentiment analysis | LREC | 51.0 | conference | International Conference on Language Resources and Evaluation | Saif M. Mohammad,Mohammad Salameh,Svetlana Kiritchenko | National Research Council Canada,, | Existing Arabic sentiment lexicons have low coverage―with only a few thousand entries. In this paper, we present several large sentiment lexicons that were automatically generated using two different methods: (1) by using distant supervision techniques on Arabic tweets, and (2) by translating English sentiment lexicons into Arabic using a freely available statistical machine translation system. We compare the usefulness of new and old sentiment lexicons in the downstream application of sentence-level sentiment analysis. Our baseline sentiment analysis system uses numerous surface form features. Nonetheless, the system benefits from using additional features drawn from sentiment lexicons. The best result is obtained using the automatically generated Dialectal Hashtag Lexicon and the Arabic translations of the NRC Emotion Lexicon (accuracy of 66.6%). Finally, we describe a qualitative study of the automatic translations of English sentiment lexicons into Arabic, which shows that about 88% of the automatically translated entries are valid for English as well. Close to 10% of the invalid entries are caused by gross mistranslations, close to 40% by translations into a related word, and about 50% by differences in how the word is used in Arabic. | Zaid Alyafeai |
Arabic Flood Twitter Dataset | [] | https://huggingface.co/datasets/arbml/twitter_flood_detection | https://github.com/alaa-a-a/Arabic-Twitter-Corpus-for-Flood-Detection | unknown | 2,019 | ar | mixed | social media | text | crawling and annotation(other) | It includes 4,037 human-labelled Arabic Twitter messages for four high-risk flood events that occurred in 2018 | 4,037 | sentences | Medium | Multiple institutions | nan | Crisis Detection from Arabic Tweets
| https://aclanthology.org/W19-5609.pdf | Arab | No | GitHub | Free | nan | No | crisis detection | WACL | 5.0 | workshop | Workshop on Arabic Corpus Linguistics | Alaa Alharbi,Mark G. Lee | Taibah University;University of Birmingham, | Social media (SM) platforms such as Twitter offer a rich source of real-time information about crises from which useful information can be extracted to support situational
awareness. The task of automatically identifying SM messages related to a specific event
poses many challenges, including processing
large volumes of short, noisy data in real time.
This paper explored the problem of extracting crisis-related messages from Arabic Twitter data. We focused on high-risk floods as
they are one of the main hazards in the Middle East. In this work, we presented a goldstandard Arabic Twitter corpus for four highrisk floods that occurred in 2018. Using the
annotated dataset, we investigated the performance of different classical machine learning
(ML) and deep neural network (DNN) classifiers. The results showed that deep learning is
promising in identifying flood-related posts | Zaid Alyafeai |
Arabic Textual Entailment Dataset | [] | nan | http://www.cs.man.ac.uk/~ramsay/ArabicTE/ | unknown | 2,013 | ar | ar-MSA: (Arabic (Modern Standard Arabic)) | news articles | text | manual curation | This dataset contains 600 Arabic premise:hypothesis pairs. | 600 | sentences | Low | Multiple institutions | nan | Natural Language Inference for Arabic Using Extended Tree
Edit Distance with Subtrees | https://arxiv.org/ftp/arxiv/papers/1402/1402.0578.pdf | Latn | No | other | Free | nan | No | natural language inference | JAIR | 21.0 | journal | Journal of Artificial Intelligence Research | Maytham Alabbas,A. Ramsay | , | Many natural language processing (NLP) applications require the computation of similarities between pairs of syntactic or semantic trees. Many researchers have used tree edit distance for this task, but this technique suffers from the drawback that it deals with single node operations only. We have extended the standard tree edit distance algorithm to deal with subtree transformation operations as well as single nodes. The extended algorithm with subtree operations, TED+ST, is more effective and flexible than the standard algorithm, especially for applications that pay attention to relations among nodes (e.g. in linguistic trees, deleting a modifier subtree should be cheaper than the sum of deleting its components individually). We describe the use of TED+ST for checking entailment between two Arabic text snippets. The preliminary results of using TED+ST were encouraging when compared with two string-based approaches and with the standard algorithm. | Zaid Alyafeai |