Back to home
All Datasets 🏷
All datasets from our datasets repository and community bucket.
Also check out the list of supported Metrics 📉.
383 results
ADE-Corpus-V2 Dataset: Adverse Drug Reaction Data. This is a dataset for Classification if a sentence is ADE-related (True) or not (False) and Relation Extraction between Adverse Drug Event and Drug. DRUG-AE.rel provides relations between drugs and adverse effects. DRUG-DOSE.rel provides relations between drugs and dosages. ADE-NEG.txt provides all sentences in the ADE corpus that DO NOT contain any drug-related adverse effects.
AdversarialQA is a Reading Comprehension dataset, consisting of questions posed by crowdworkers on a set of Wikipedia articles using an adversarial model-in-the-loop. We use three different models; BiDAF (Seo et al., 2016), BERT-Large (Devlin et al., 2018), and RoBERTa-Large (Liu et al., 2019) in the annotation loop and construct three datasets; D(BiDAF), D(BERT), and D(RoBERTa), each with 10,000 training examples, 1,000 validation, and 1,000 test examples. The adversarial human annotation paradigm ensures that these datasets consist of questions that current state-of-the-art models (at least the ones used as adversaries in the annotation loop) find challenging.
AirDialogue, is a large dataset that contains 402,038 goal-oriented conversations. To collect this dataset, we create a contextgenerator which provides travel and flight restrictions. Then the human annotators are asked to play the role of a customer or an agent and interact with the goal of successfully booking a trip given the restrictions.
Allegro Reviews is a sentiment analysis dataset, consisting of 11,588 product reviews written in Polish and extracted from - a popular e-commerce marketplace. Each review contains at least 50 words and has a rating on a scale from one (negative review) to five (positive review). We recommend using the provided train/dev/test split. The ratings for the test set reviews are kept hidden. You can evaluate your model using the online evaluation tool available on
Allocine Dataset: A Large-Scale French Movie Reviews Dataset. This is a dataset for binary sentiment classification, made of user reviews scraped from It contains 100k positive and 100k negative reviews divided into 3 balanced splits: train (160k reviews), val (20k) and test (20k).
The ALT project aims to advance the state-of-the-art Asian natural language processing (NLP) techniques through the open collaboration for developing and using ALT. It was first conducted by NICT and UCSY as described in Ye Kyaw Thu, Win Pa Pa, Masao Utiyama, Andrew Finch and Eiichiro Sumita (2016). Then, it was developed under ASEAN IVO as described in this Web page. The process of building ALT began with sampling about 20,000 sentences from English Wikinews, and then these sentences were translated into the other languages. ALT now has 13 languages: Bengali, English, Filipino, Hindi, Bahasa Indonesia, Japanese, Khmer, Lao, Malay, Myanmar (Burmese), Thai, Vietnamese, Chinese (Simplified Chinese).
The Amazon reviews dataset consists of reviews from amazon. The data span a period of 18 years, including ~35 million reviews up to March 2013. Reviews include product and user information, ratings, and a plaintext review.
We provide an Amazon product reviews dataset for multilingual text classification. The dataset contains reviews in English, Japanese, German, French, Chinese and Spanish, collected between November 1, 2015 and November 1, 2019. Each record in the dataset contains the review text, the review title, the star rating, an anonymized reviewer ID, an anonymized product ID and the coarse-grained product category (e.g. ‘books’, ‘appliances’, etc.) The corpus is balanced across stars, so each star rating constitutes 20% of the reviews in each language. For each language, there are 200,000, 5,000 and 5,000 reviews in the training, development and test sets respectively. The maximum number of reviews per reviewer is 20 and the maximum number of reviews per product is 20. All reviews are truncated after 2,000 characters, and all reviews are at least 20 characters long. Note that the language of a review does not necessarily match the language of its marketplace (e.g. reviews from are primarily written in German, but could also be written in English, etc.). For this reason, we applied a language detection algorithm based on the work in Bojanowski et al. (2017) to determine the language of the review text and we removed reviews that were not written in the expected language.
AmbigNQ, a dataset covering 14,042 questions from NQ-open, an existing open-domain QA benchmark. We find that over half of the questions in NQ-open are ambiguous. The types of ambiguity are diverse and sometimes subtle, many of which are only apparent after examining evidence provided by a very large text corpus. AMBIGNQ, a dataset with 14,042 annotations on NQ-OPEN questions containing diverse types of ambiguity. We provide two distributions of our new dataset AmbigNQ: a full version with all annotation metadata and a light version with only inputs and outputs.
Chinese word segmentation (CWS) trained from open source corpus faces dramatic performance drop when dealing with domain text, especially for a domain with lots of special terms and diverse writing styles, such as the biomedical domain. However, building domain-specific CWS requires extremely high annotation cost. In this paper, we propose an approach by exploiting domain-invariant knowledge from high resource to low resource domains. Extensive experiments show that our mode achieves consistently higher accuracy than the single-task CWS and other transfer learning baselines, especially when there is a large disparity between source and target domains. This dataset is the accompanied medical Chinese word segmentation (CWS) dataset. The tags are in BIES scheme. For more details see
It is a large dataset of Android applications belonging to 23 differentapps categories, which provides an overview of the types of feedback users report on the apps and documents the evolution of the related code metrics. The dataset contains about 395 applications of the F-Droid repository, including around 600 versions, 280,000 user reviews (extracted with specific text mining approaches)
A large-scale dataset consisting of approximately 100,000 algebraic word problems. The solution to each question is explained step-by-step using natural language. This data is used to train a program generation model that learns to generate the explanation, while generating the program that solves the question.
ArCOV-19 is an Arabic COVID-19 Twitter dataset that covers the period from 27th of January till 30th of April 2020. ArCOV-19 is designed to enable research under several domains including natural language processing, information retrieval, and social computing, among others
Abu El-Khair Corpus is an Arabic text corpus, that includes more than five million newspaper articles. It contains over a billion and a half words in total, out of which, there are about three million unique words. The corpus is 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.
ASSET is a dataset for evaluating Sentence Simplification systems with multiple rewriting transformations, as described in "ASSET: A Dataset for Tuning and Evaluation of Sentence Simplification Models with Multiple Rewriting Transformations". The corpus is composed of 2000 validation and 359 test original sentences that were each simplified 10 times by different annotators. The corpus also contains human judgments of meaning preservation, fluency and simplicity for the outputs of several automatic text simplification systems.
The ASSIN (Avaliação de Similaridade Semântica e INferência textual) corpus is a corpus annotated with pairs of sentences written in Portuguese that is suitable for the exploration of textual entailment and paraphrasing classifiers. The corpus contains pairs of sentences extracted from news articles written in European Portuguese (EP) and Brazilian Portuguese (BP), obtained from Google News Portugal and Brazil, respectively. To create the corpus, the authors started by collecting a set of news articles describing the same event (one news article from Google News Portugal and another from Google News Brazil) from Google News. Then, they employed Latent Dirichlet Allocation (LDA) models to retrieve pairs of similar sentences between sets of news articles that were grouped together around the same topic. For that, two LDA models were trained (for EP and for BP) on external and large-scale collections of unannotated news articles from Portuguese and Brazilian news providers, respectively. Then, the authors defined a lower and upper threshold for the sentence similarity score of the retrieved pairs of sentences, taking into account that high similarity scores correspond to sentences that contain almost the same content (paraphrase candidates), and low similarity scores correspond to sentences that are very different in content from each other (no-relation candidates). From the collection of pairs of sentences obtained at this stage, the authors performed some manual grammatical corrections and discarded some of the pairs wrongly retrieved. Furthermore, from a preliminary analysis made to the retrieved sentence pairs the authors noticed that the number of contradictions retrieved during the previous stage was very low. Additionally, they also noticed that event though paraphrases are not very frequent, they occur with some frequency in news articles. Consequently, in contrast with the majority of the currently available corpora for other languages, which consider as labels “neutral”, “entailment” and “contradiction” for the task of RTE, the authors of the ASSIN corpus decided to use as labels “none”, “entailment” and “paraphrase”. Finally, the manual annotation of pairs of sentences was performed by human annotators. At least four annotators were randomly selected to annotate each pair of sentences, which is done in two steps: (i) assigning a semantic similarity label (a score between 1 and 5, from unrelated to very similar); and (ii) providing an entailment label (one sentence entails the other, sentences are paraphrases, or no relation). Sentence pairs where at least three annotators do not agree on the entailment label were considered controversial and thus discarded from the gold standard annotations. The full dataset has 10,000 sentence pairs, half of which in Brazilian Portuguese and half in European Portuguese. Either language variant has 2,500 pairs for training, 500 for validation and 2,000 for testing.
The ASSIN 2 corpus is composed of rather simple sentences. Following the procedures of SemEval 2014 Task 1. The training and validation data are composed, respectively, of 6,500 and 500 sentence pairs in Brazilian Portuguese, annotated for entailment and semantic similarity. Semantic similarity values range from 1 to 5, and text entailment classes are either entailment or none. The test data are composed of approximately 3,000 sentence pairs with the same annotation. All data were manually annotated.
This dataset provides the template sentences and relationships defined in the ATOMIC common sense dataset. There are three splits - train, test, and dev. From the authors. Disclaimer/Content warning: the events in atomic have been automatically extracted from blogs, stories and books written at various times. The events might depict violent or problematic actions, which we left in the corpus for the sake of learning the (probably negative but still important) commonsense implications associated with the events. We removed a small set of truly out-dated events, but might have missed some so please email us ( if you have any concerns.
Multilingual information access is stipulated in the South African constitution. In practise, this is hampered by a lack of resources and capacity to perform the large volumes of translation work required to realise multilingual information access. One of the aims of the Autshumato project is to develop machine translation systems for three South African languages pairs.
Nineteen teams presented results for the Gene Mention Task at the BioCreative II Workshop. In this task participants designed systems to identify substrings in sentences corresponding to gene name mentions. A variety of different methods were used and the results varied with a highest achieved F1 score of 0.8721. Here we present brief descriptions of all the methods used and a statistical analysis of the results. We also demonstrate that, by combining the results from all submissions, an F score of 0.9066 is feasible, and furthermore that the best result makes use of the lowest scoring submissions. For more details, see: The original dataset can be downloaded from: This dataset has been converted to CoNLL format for NER using the following tool:
`best2009` is a Thai word-tokenization dataset from encyclopedia, novels, news and articles by [NECTEC]( (148,995/2,252 lines of train/test). It was created for [BEST 2010: Word Tokenization Competition]( The test set answers are not provided publicly.
A parallel news corpus in Turkish, Kurdish and English. Bianet collects 3,214 Turkish articles with their sentence-aligned Kurdish or English translations from the Bianet online newspaper. 3 languages, 3 bitexts total number of files: 6 total number of tokens: 2.25M total number of sentence fragments: 0.14M
This is a multilingual parallel corpus created from translations of the Bible compiled by Christos Christodoulopoulos and Mark Steedman. 102 languages, 5,148 bitexts total number of files: 107 total number of tokens: 56.43M total number of sentence fragments: 2.84M
BIGPATENT, consisting of 1.3 million records of U.S. patent documents along with human written abstractive summaries. Each US patent application is filed under a Cooperative Patent Classification (CPC) code. There are nine such classification categories: A (Human Necessities), B (Performing Operations; Transporting), C (Chemistry; Metallurgy), D (Textiles; Paper), E (Fixed Constructions), F (Mechanical Engineering; Lightning; Heating; Weapons; Blasting), G (Physics), H (Electricity), and Y (General tagging of new or cross-sectional technology) There are two features: - description: detailed description of patent. - abstract: Patent abastract.
This dataset was curated from the Bing search logs (desktop users only) over the period of Jan 1st, 2020 – (Current Month - 1). Only searches that were issued many times by multiple users were included. The dataset includes queries from all over the world that had an intent related to the Coronavirus or Covid-19. In some cases this intent is explicit in the query itself (e.g., “Coronavirus updates Seattle”), in other cases it is implicit , e.g. “Shelter in place”. The implicit intent of search queries (e.g., “Toilet paper”) was extracted using random walks on the click graph as outlined in this paper by Microsoft Research. All personal data were removed.
The Bengali Hate Speech Dataset is a collection of Bengali articles collected from Bengali news articles, news dump of Bengali TV channels, books, blogs, and social media. Emphasis was placed on Facebook pages and newspaper sources because they attract close to 50 million followers and is a common source of opinions and hate speech. The raw text corpus contains 250 million articles and the full dataset is being prepared for release. This is a subset of the full dataset. This dataset was prepared for hate-speech text classification benchmark on Bengali, an under-resourced language.
The BrWaC (Brazilian Portuguese Web as Corpus) is a large corpus constructed following the Wacky framework, which was made public for research purposes. The current corpus version, released in January 2017, is composed by 3.53 million documents, 2.68 billion tokens and 5.79 million types. Please note that this resource is available solely for academic research purposes, and you agreed not to use it for any commercial applications. Manually download at
This is the Business Scene Dialogue (BSD) dataset, a Japanese-English parallel corpus containing written conversations in various business scenarios. The dataset was constructed in 3 steps: 1) selecting business scenes, 2) writing monolingual conversation scenarios according to the selected scenes, and 3) translating the scenarios into the other language. Half of the monolingual scenarios were written in Japanese and the other half were written in English. Fields: - id: dialogue identifier - no: sentence pair number within a dialogue - en_speaker: speaker name in English - ja_speaker: speaker name in Japanese - en_sentence: sentence in English - ja_sentence: sentence in Japanese - original_language: language in which monolingual scenario was written - tag: scenario - title: scenario title
The Bosnian web corpus bsWaC was built by crawling the .ba top-level domain in 2014. The corpus was near-deduplicated on paragraph level, normalised via diacritic restoration, morphosyntactically annotated and lemmatised. The corpus is shuffled by paragraphs. Each paragraph contains metadata on the URL, domain and language identification (Bosnian vs. Croatian vs. Serbian). Version 1.0 of this corpus is described in Version 1.1 contains newer and better linguistic annotations.
Machine reading comprehension tasks require a machine reader to answer questions relevant to the given document. In this paper, we present the first free-form multiple-Choice Chinese machine reading Comprehension dataset (C^3), containing 13,369 documents (dialogues or more formally written mixed-genre texts) and their associated 19,577 multiple-choice free-form questions collected from Chinese-as-a-second-language examinations. We present a comprehensive analysis of the prior knowledge (i.e., linguistic, domain-specific, and general world knowledge) needed for these real-world problems. We implement rule-based and popular neural methods and find that there is still a significant performance gap between the best performing model (68.5%) and human readers (96.0%), especially on problems that require prior knowledge. We further study the effects of distractor plausibility and data augmentation based on translated relevant datasets for English on model performance. We expect C^3 to present great challenges to existing systems as answering 86.8% of questions requires both knowledge within and beyond the accompanying document, and we hope that C^3 can serve as a platform to study how to leverage various kinds of prior knowledge to better understand a given written or orally oriented text.
In this paper, we introduce Chinese AI and Law challenge dataset (CAIL2018), the first large-scale Chinese legal dataset for judgment prediction. CAIL contains more than 2.6 million criminal cases published by the Supreme People's Court of China, which are several times larger than other datasets in existing works on judgment prediction. Moreover, the annotations of judgment results are more detailed and rich. It consists of applicable law articles, charges, and prison terms, which are expected to be inferred according to the fact descriptions of cases. For comparison, we implement several conventional text classification baselines for judgment prediction and experimental results show that it is still a challenge for current models to predict the judgment results of legal cases, especially on prison terms. To help the researchers make improvements on legal judgment prediction.
A parallel corpus of theses and dissertations abstracts in English and Portuguese were collected from the CAPES website (Coordenação de Aperfeiçoamento de Pessoal de Nível Superior) - Brazil. The corpus is sentence aligned for all language pairs. Approximately 240,000 documents were collected and aligned using the Hunalign algorithm.
This dataset contains two corpora in Spanish and Catalan that consist of annotated Twitter messages for automatic stance detection. The data was collected over 12 days during February and March of 2019 from tweets posted in Barcelona, and during September of 2018 from tweets posted in the town of Terrassa, Catalonia. Each corpus is annotated with three classes: AGAINST, FAVOR and NEUTRAL, which express the stance towards the target - independence of Catalonia.
This corpus is an attempt to recreate the dataset used for training XLM-R. This corpus comprises of monolingual data for 100+ languages and also includes data for romanized languages (indicated by *_rom). This was constructed using the urls and paragraph indices provided by the CC-Net repository by processing January-December 2018 Commoncrawl snapshots. Each file comprises of documents separated by double-newlines and paragraphs within the same document separated by a newline. The data is generated using the open source CC-Net repository. No claims of intellectual property are made on the work of preparation of the corpus.
Polish CDSCorpus consists of 10K Polish sentence pairs which are human-annotated for semantic relatedness and entailment. The dataset may be used for the evaluation of compositional distributional semantics models of Polish. The dataset was presented at ACL 2017. Please refer to the Wróblewska and Krasnowska-Kieraś (2017) for a detailed description of the resource.
ChrEn is a Cherokee-English parallel dataset to facilitate machine translation research between Cherokee and English. ChrEn is extremely low-resource contains 14k sentence pairs in total, split in ways that facilitate both in-domain and out-of-domain evaluation. ChrEn also contains 5k Cherokee monolingual data to enable semi-supervised learning.
The Circa (meaning ‘approximately’) dataset aims to help machine learning systems to solve the problem of interpreting indirect answers to polar questions. The dataset contains pairs of yes/no questions and indirect answers, together with annotations for the interpretation of the answer. The data is collected in 10 different social conversational situations (eg. food preferences of a friend). NOTE: There might be missing labels in the dataset and we have replaced them with -1. The original dataset contains no train/dev/test splits.
A dataset adopting the FEVER methodology that consists of 1,535 real-world claims regarding climate-change collected on the internet. Each claim is accompanied by five manually annotated evidence sentences retrieved from the English Wikipedia that support, refute or do not give enough information to validate the claim totalling in 7,675 claim-evidence pairs. The dataset features challenging claims that relate multiple facets and disputed cases of claims where both supporting and refuting evidence are present.
This dataset is for evaluating the performance of intent classification systems in the presence of "out-of-scope" queries. By "out-of-scope", we mean queries that do not fall into any of the system-supported intent classes. Most datasets include only data that is "in-scope". Our dataset includes both in-scope and out-of-scope data. You might also know the term "out-of-scope" by other terms, including "out-of-domain" or "out-of-distribution".
CNN/DailyMail non-anonymized summarization dataset. There are two features: - article: text of news article, used as the document to be summarized - highlights: joined text of highlights with <s> and </s> around each highlight, which is the target summary
A dataset consisting of 502 English dialogs with 12,000 annotated utterances between a user and an assistant discussing movie preferences in natural language. It was collected using a Wizard-of-Oz methodology between two paid crowd-workers, where one worker plays the role of an 'assistant', while the other plays the role of a 'user'. The 'assistant' elicits the 'user’s' preferences about movies following a Coached Conversational Preference Elicitation (CCPE) method. The assistant asks questions designed to minimize the bias in the terminology the 'user' employs to convey his or her preferences as much as possible, and to obtain these preferences in natural language. Each dialog is annotated with entity mentions, preferences expressed about entities, descriptions of entities provided, and other statements of entities.
The COmmonsense Dataset Adversarially-authored by Humans (CODAH) is an evaluation set for commonsense question-answering in the sentence completion style of SWAG. As opposed to other automatically generated NLI datasets, CODAH is adversarially constructed by humans who can view feedback from a pre-trained model and use this information to design challenging commonsense questions. Our experimental results show that CODAH questions present a complementary extension to the SWAG dataset, testing additional modes of common sense.
CodeSearchNet corpus contains about 6 million functions from open-source code spanning six programming languages (Go, Java, JavaScript, PHP, Python, and Ruby). The CodeSearchNet Corpus also contains automatically generated query-like natural language for 2 million functions, obtained from mechanically scraping and preprocessing associated function documentation.
This dataset is designed to provide training data for common sense relationships pulls together from various sources. The dataset is multi-lingual. See langauge codes and language info here: This dataset provides an interface for the conceptnet5 csv file, and some (but not all) of the raw text data used to build conceptnet5: omcsnet_sentences_free.txt, and omcsnet_sentences_more.txt. One use of this dataset would be to learn to extract the conceptnet relationship from the omcsnet sentences. Conceptnet5 has 34,074,917 relationships. Of those relationships, there are 2,176,099 surface text sentences related to those 2M entries. omcsnet_sentences_free has 898,161 lines. omcsnet_sentences_more has 2,001,736 lines. Original downloads are available here For more information, see: The omcsnet data comes with the following warning from the authors of the above site: Remember: this data comes from various forms of crowdsourcing. Sentences in these files are not necessarily true, useful, or appropriate.
Named entities are phrases that contain the names of persons, organizations, locations, times and quantities. Example: [PER Wolff] , currently a journalist in [LOC Argentina] , played with [PER Del Bosque] in the final years of the seventies in [ORG Real Madrid] . The shared task of CoNLL-2002 concerns language-independent named entity recognition. We will concentrate on four types of named entities: persons, locations, organizations and names of miscellaneous entities that do not belong to the previous three groups. The participants of the shared task will be offered training and test data for at least two languages. They will use the data for developing a named-entity recognition system that includes a machine learning component. Information sources other than the training data may be used in this shared task. We are especially interested in methods that can use additional unannotated data for improving their performance (for example co-training). The train/validation/test sets are available in Spanish and Dutch. For more details see and
ConvAI is a dataset of human-to-bot conversations labelled for quality. This data can be used to train a metric for evaluating dialogue systems. Moreover, it can be used in the development of chatbots themselves: it contains the information on the quality of utterances and entire dialogues, that can guide a dialogue system in search of better answers.
ConvAI is a dataset of human-to-bot conversations labelled for quality. This data can be used to train a metric for evaluating dialogue systems. Moreover, it can be used in the development of chatbots themselves: it contains the information on the quality of utterances and entire dialogues, that can guide a dialogue system in search of better answers.
The Conv AI 3 challenge is organized as part of the Search-oriented Conversational AI (SCAI) EMNLP workshop in 2020. The main aim of the conversational systems is to return an appropriate answer in response to the user requests. However, some user requests might be ambiguous. In Information Retrieval (IR) settings such a situation is handled mainly through the diversification of search result page. It is however much more challenging in dialogue settings. Hence, we aim to study the following situation for dialogue settings: - a user is asking an ambiguous question (where ambiguous question is a question to which one can return > 1 possible answers) - the system must identify that the question is ambiguous, and, instead of trying to answer it directly, ask a good clarifying question.
The COrpus of Urdu News TExt Reuse (COUNTER) corpus contains 1200 documents with real examples of text reuse from the field of journalism. It has been manually annotated at document level with three levels of reuse: wholly derived, partially derived and non derived.
53,640 Japanese tweets with annotation if a tweet is related to COVID-19 or not. The annotation is by majority decision by 5 - 10 crowd workers. Target tweets include "COVID" or "コロナ". The period of the tweets is from around January 2020 to around June 2020. The original tweets are not contained. Please use Twitter API to get them, for example.
We study negotiation dialogues where two agents, a buyer and a seller, negotiate over the price of an time for sale. We collected a dataset of more than 6K negotiation dialogues over multiple categories of products scraped from Craigslist. Our goal is to develop an agent that negotiates with humans through such conversations. The challenge is to handle both the negotiation strategy and the rich language for bargaining.
Corpus of domain names scraped from Common Crawl and manually annotated to add word boundaries (e.g. "commoncrawl" to "common crawl"). Breaking domain names such as "openresearch" into component words "open" and "research" is important for applications such as Text-to-Speech synthesis and web search. Common Crawl is an open repository of web crawl data that can be accessed and analyzed by anyone. Specifically, we scraped the plaintext (WET) extracts for domain names from URLs that contained diverse letter casing (e.g. "OpenBSD"). Although in the previous example, segmentation is trivial using letter casing, this was not always the case (e.g. "NASA"), so we had to manually annotate the data. The dataset is stored as plaintext file where each line is an example of space separated segments of a domain name. The examples are stored in their original letter casing, but harder and more interesting examples can be generated by lowercasing the input first.
This dataset contains 14K dialogs (181K utterances) where users and assistants converse about geographic topics like geopolitical entities and locations. This dataset is annotated with pre-existing user knowledge, message-level dialog acts, grounding to Wikipedia, and user reactions to messages.
The dataset consists of 9008 sentences that are labelled with fine-grained polarity in the range from -2 to 2 (negative to postive). The quality of the fine-grained is not cross validated and is therefore subject to uncertainties; however, the simple polarity has been cross validated and therefore is considered to be more correct.
The DBpedia ontology classification dataset is constructed by picking 14 non-overlapping classes from DBpedia 2014. They are listed in classes.txt. From each of thse 14 ontology classes, we randomly choose 40,000 training samples and 5,000 testing samples. Therefore, the total size of the training dataset is 560,000 and testing dataset 70,000. There are 3 columns in the dataset (same for train and test splits), corresponding to class index (1 to 14), title and content. The title and content are escaped using double quotes ("), and any internal double quote is escaped by 2 double quotes (""). There are no new lines in title or content.
The Dutch Book Review Dataset (DBRD) contains over 110k book reviews of which 22k have associated binary sentiment polarity labels. It is intended as a benchmark for sentiment classification in Dutch and created due to a lack of annotated datasets in Dutch that are suitable for this task.
Benchmark dataset for low-resource multiclass classification, with 4,015 training, 500 testing, and 500 validation examples, each labeled as part of five classes. Each sample can be a part of multiple classes. Collected as tweets.
DialogRE is the first human-annotated dialogue based relation extraction (RE) dataset aiming to support the prediction of relation(s) between two arguments that appear in a dialogue. The dataset annotates all occurrences of 36 possible relation types that exist between pairs of arguments in the 1,788 dialogues originating from the complete transcripts of Friends.
This dataset contains 30,000 messages drawn from events including an earthquake in Haiti in 2010, an earthquake in Chile in 2010, floods in Pakistan in 2010, super-storm Sandy in the U.S.A. in 2012, and news articles spanning a large number of years and 100s of different disasters. The data has been encoded with 36 different categories related to disaster response and has been stripped of messages with sensitive information in their entirety. Upon release, this is the featured dataset of a new Udacity course on Data Science and the AI4ALL summer school and is especially utile for text analytics and natural language processing (NLP) tasks and models. The input data in this job contains thousands of untranslated disaster-related messages and their English translations.
Doc2dial is dataset of goal-oriented dialogues that are grounded in the associated documents. It includes over 4500 annotated conversations with an average of 14 turns that are grounded in over 450 documents from four domains. Compared to the prior document-grounded dialogue datasets this dataset covers a variety of dialogue scenes in information-seeking conversations.
DREAM is a multiple-choice Dialogue-based REAding comprehension exaMination dataset. In contrast to existing reading comprehension datasets, DREAM is the first to focus on in-depth multi-turn multi-party dialogue understanding.
The dataset contains around 271,342 tweets. The tweets are filtered via the official Twitter API to contain tweets in Dutch language or by users who have specified their location information within Netherlands geographical boundaries. Using natural language processing we have classified the tweets for their HISCO codes. If the user has provided their location within Dutch boundaries, we have also classified them to their respective provinces The objective of this dataset is to make research data available publicly in a FAIR (Findable, Accessible, Interoperable, Reusable) way. Twitter's Terms of Service Licensed under Attribution-NonCommercial 4.0 International (CC BY-NC 4.0) (2020-10-27)
The Did You Know (pol. Czy wiesz?) dataset consists of human-annotated question-answer pairs. The task is to predict if the answer is correct. We chose the negatives which have the largest token overlap with a question.
The E2E dataset is used for training end-to-end, data-driven natural language generation systems in the restaurant domain, which is ten times bigger than existing, frequently used datasets in this area. The E2E dataset poses new challenges: (1) its human reference texts show more lexical richness and syntactic variation, including discourse phenomena; (2) generating from this set requires content selection. As such, learning from this dataset promises more natural, varied and less template-like system utterances. E2E is released in the following paper where you can find more details and baseline results:
An update release of E2E NLG Challenge data with cleaned MRs and scripts, accompanying the following paper: Ondřej Dušek, David M. Howcroft, and Verena Rieser (2019): Semantic Noise Matters for Neural Natural Language Generation. In INLG, Tokyo, Japan.
Original source: Website and documentatuion from the European Central Bank, compiled and made available by Alberto Simoes (thank you very much!) 19 languages, 170 bitexts total number of files: 340 total number of tokens: 757.37M total number of sentence fragments: 30.55M
EiTB-ParCC: Parallel Corpus of Comparable News. A Basque-Spanish parallel corpus provided by Vicomtech (, extracted from comparable news produced by the Basque public broadcasting group Euskal Irrati Telebista.
This is a parallel corpus made out of PDF documents from the European Medicines Agency. All files are automatically converted from PDF to plain text using pdftotext with the command line arguments -layout -nopgbrk -eol unix. There are some known problems with tables and multi-column layouts - some of them are fixed in the current version. source: 22 languages, 231 bitexts total number of files: 41,957 total number of tokens: 311.65M total number of sentence fragments: 26.51M
ETHOS: onlinE haTe speecH detectiOn dataSet. This repository contains a dataset for hate speech detection on social media platforms, called Ethos. There are two variations of the dataset: Ethos_Dataset_Binary: contains 998 comments in the dataset alongside with a label about hate speech presence or absence. 565 of them do not contain hate speech, while the rest of them, 433, contain. Ethos_Dataset_Multi_Label: which contains 8 labels for the 433 comments with hate speech content. These labels are violence (if it incites (1) or not (0) violence), directed_vs_general (if it is directed to a person (1) or a group (0)), and 6 labels about the category of hate speech like, gender, race, national_origin, disability, religion and sexual_orientation.
The corpora comprise of files per data provider that are encoded in the IOB format (Ramshaw & Marcus, 1995). The IOB format is a simple text chunking format that divides texts into single tokens per line, and, separated by a whitespace, tags to mark named entities. The most commonly used categories for tags are PER (person), LOC (location) and ORG (organization). To mark named entities that span multiple tokens, the tags have a prefix of either B- (beginning of named entity) or I- (inside of named entity). O (outside of named entity) tags are used to mark tokens that are not a named entity.
In October 2012, the European Union's (EU) Directorate General for Education and Culture ( DG EAC) released a translation memory (TM), i.e. a collection of sentences and their professionally produced translations, in twenty-six languages. This resource bears the name EAC Translation Memory, short EAC-TM. EAC-TM covers up to 26 languages: 22 official languages of the EU (all except Irish) plus Icelandic, Croatian, Norwegian and Turkish. EAC-TM thus contains translations from English into the following 25 languages: Bulgarian, Czech, Danish, Dutch, Estonian, German, Greek, Finnish, French, Croatian, Hungarian, Icelandic, Italian, Latvian, Lithuanian, Maltese, Norwegian, Polish, Portuguese, Romanian, Slovak, Slovenian, Spanish, Swedish and Turkish. All documents and sentences were originally written in English (source language is English) and then translated into the other languages. The texts were translated by staff of the National Agencies of the Lifelong Learning and Youth in Action programmes. They are typically professionals in the field of education/youth and EU programmes. They are thus not professional translators, but they are normally native speakers of the target language.
In October 2012, the European Union (EU) agency 'European Centre for Disease Prevention and Control' (ECDC) released a translation memory (TM), i.e. a collection of sentences and their professionally produced translations, in twenty-five languages. This resource bears the name EAC Translation Memory, short EAC-TM. ECDC-TM covers 25 languages: the 23 official languages of the EU plus Norwegian (Norsk) and Icelandic. ECDC-TM was created by translating from English into the following 24 languages: Bulgarian, Czech, Danish, Dutch, English, Estonian, Gaelige (Irish), German, Greek, Finnish, French, Hungarian, Icelandic, Italian, Latvian, Lithuanian, Maltese, Norwegian (NOrsk), Polish, Portuguese, Romanian, Slovak, Slovenian, Spanish and Swedish. All documents and sentences were thus originally written in English. They were then translated into the other languages by professional translators from the Translation Centre CdT in Luxembourg.
Data and code from our "Inferring Which Medical Treatments Work from Reports of Clinical Trials", NAACL 2019. This work concerns inferring the results reported in clinical trials from text. The dataset consists of biomedical articles describing randomized control trials (RCTs) that compare multiple treatments. Each of these articles will have multiple questions, or 'prompts' associated with them. These prompts will ask about the relationship between an intervention and comparator with respect to an outcome, as reported in the trial. For example, a prompt may ask about the reported effects of aspirin as compared to placebo on the duration of headaches. For the sake of this task, we assume that a particular article will report that the intervention of interest either significantly increased, significantly decreased or had significant effect on the outcome, relative to the comparator. The dataset could be used for automatic data extraction of the results of a given RCT. This would enable readers to discover the effectiveness of different treatments without needing to read the paper.
EXAMS is a benchmark dataset for multilingual and cross-lingual question answering from high school examinations. It consists of more than 24,000 high-quality high school exam questions in 16 languages, covering 8 language families and 24 school subjects from Natural Sciences and Social Sciences, among others.
A dataset to study Fake News in Portuguese, presenting a supposedly false News along with their respective fact check and classification. The data is collected from the ClaimReview, a structured data schema used by fact check agencies to share their results in search engines, enabling data collect in real time. The FACTCK.BR dataset contains 1309 claims with its corresponding label.
Fake news has become a major societal issue and a technical challenge for social media companies to identify. This content is difficult to identify because the term "fake news" covers intentionally false, deceptive stories as well as factual errors, satire, and sometimes, stories that a person just does not like. Addressing the problem requires clear definitions and examples. In this work, we present a dataset of fake news and satire stories that are hand coded, verified, and, in the case of fake news, include rebutting stories. We also include a thematic content analysis of the articles, identifying major themes that include hyperbolic support or condemnation of a gure, conspiracy theories, racist themes, and discrediting of reliable sources. In addition to releasing this dataset for research use, we analyze it and show results based on language that are promising for classification purposes. Overall, our contribution of a dataset and initial analysis are designed to support future work by fake news researchers.
Contains Farsi (Persian) datasets for Machine Learning tasks, particularly NLP. These datasets have been extracted from the RSS feed of two Farsi news agency websites: - Hamshahri - RadioFarda
The directory data contains a corpus of Finnish technology related news articles with a manually prepared named entity annotation (digitoday.2014.csv). The text material was extracted from the archives of Digitoday, a Finnish online technology news source ( The corpus consists of 953 articles (193,742 word tokens) with six named entity classes (organization, location, person, product, event, and date). The corpus is available for research purposes and can be readily used for development of NER systems for Finnish.
`generated_reviews_enth` Generated product reviews dataset for machine translation quality prediction, part of [scb-mt-en-th-2020]( `generated_reviews_enth` is created as part of [scb-mt-en-th-2020]( for machine translation task. This dataset (referred to as `generated_reviews_yn` in [scb-mt-en-th-2020]( are English product reviews generated by [CTRL](, translated by Google Translate API and annotated as accepted or rejected (`correct`) based on fluency and adequacy of the translation by human annotators. This allows it to be used for English-to-Thai translation quality esitmation (binary label), machine translation, and sentiment analysis.
The GenericsKB contains 3.4M+ generic sentences about the world, i.e., sentences expressing general truths such as "Dogs bark," and "Trees remove carbon dioxide from the atmosphere." Generics are potentially useful as a knowledge source for AI systems requiring general world knowledge. The GenericsKB is the first large-scale resource containing naturally occurring generic sentences (as opposed to extracted or crowdsourced triples), and is rich in high-quality, general, semantically complete statements. Generics were primarily extracted from three large text sources, namely the Waterloo Corpus, selected parts of Simple Wikipedia, and the ARC Corpus. A filtered, high-quality subset is also available in GenericsKB-Best, containing 1,020,868 sentences. We recommend you start with GenericsKB-Best.
GermaNER is a freely available statistical German Named Entity Tagger based on conditional random fields(CRF). The tagger is trained and evaluated on the NoSta-D Named Entity dataset, which was used in the GermEval 2014 for named entity recognition. The tagger comes close to the performance of the best (proprietary) system in the competition with 77% F-measure (this is the latest result; the one reported in the paper is 76%) test set performance on the four standard NER classes (PERson, LOCation, ORGanisation and OTHer). We describe a range of features and their influence on German NER classification and provide a comparative evaluation and some analysis of the results. The software components, the training data and all data used for feature generation are distributed under permissive licenses, thus this tagger can be used in academic and commercial settings without restrictions or fees. The tagger is available as a command-line tool and as an Apache UIMA component.
Giga-word corpus for French-English from WMT2010 collected by Chris Callison-Burch 2 languages, total number of files: 452 total number of tokens: 1.43G total number of sentence fragments: 47.55M
The GoEmotions dataset contains 58k carefully curated Reddit comments labeled for 27 emotion categories or Neutral. The emotion categories are admiration, amusement, anger, annoyance, approval, caring, confusion, curiosity, desire, disappointment, disapproval, disgust, embarrassment, excitement, fear, gratitude, grief, joy, love, nervousness, optimism, pride, realization, relief, remorse, sadness, surprise.
Strongly Generalizable Question Answering (GrailQA) is a new large-scale, high-quality dataset for question answering on knowledge bases (KBQA) on Freebase with 64,331 questions annotated with both answers and corresponding logical forms in different syntax (i.e., SPARQL, S-expression, etc.). It can be used to test three levels of generalization in KBQA: i.i.d., compositional, and zero-shot.
The dataset for the variable-misuse task, described in the ICLR 2020 paper 'Global Relational Models of Source Code' [] This is the public version of the dataset used in that paper. The original, used to produce the graphs in the paper, could not be open-sourced due to licensing issues. See the public associated code repository [] for results produced from this dataset. This dataset was generated synthetically from the corpus of Python code in the ETH Py150 Open dataset [].
This dataset contains 93700 hotel reviews in Arabic language.The hotel reviews were collected from website during June/July 2016.The reviews are expressed in Modern Standard Arabic as well as dialectal Arabic.The following table summarize some tatistics on the HARD Dataset.
The HAREM is a Portuguese language corpus commonly used for Named Entity Recognition tasks. It includes about 93k words, from 129 different texts, from several genres, and language varieties. The split of this dataset version follows the division made by [1], where 7% HAREM documents are the validation set and the miniHAREM corpus (with about 65k words) is the test set. There are two versions of the dataset set, a version that has a total of 10 different named entity classes (Person, Organization, Location, Value, Date, Title, Thing, Event, Abstraction, and Other) and a "selective" version with only 5 classes (Person, Organization, Location, Value, and Date). It's important to note that the original version of the HAREM dataset has 2 levels of NER details, namely "Category" and "Sub-type". The dataset version processed here ONLY USE the "Category" level of the original dataset. [1] Souza, Fábio, Rodrigo Nogueira, and Roberto Lotufo. "BERTimbau: Pretrained BERT Models for Brazilian Portuguese." Brazilian Conference on Intelligent Systems. Springer, Cham, 2020.
These files contain text extracted from Stormfront, a white supremacist forum. A random set of forums posts have been sampled from several subforums and split into sentences. Those sentences have been manually labelled as containing hate speech or not, according to certain annotation guidelines.
Hatexplain is the first benchmark hate speech dataset covering multiple aspects of the issue. Each post in the dataset is annotated from three different perspectives: the basic, commonly used 3-class classification (i.e., hate, offensive or normal), the target community (i.e., the community that has been the victim of hate speech/offensive speech in the post), and the rationales, i.e., the portions of the post on which their labelling decision (as hate, offensive or normal) is based.
The Hausa VOA NER dataset is a labeled dataset for named entity recognition in Hausa. The texts were obtained from Hausa Voice of America News articles . We concentrate on four types of named entities: persons [PER], locations [LOC], organizations [ORG], and dates & time [DATE]. The Hausa VOA NER data files contain 2 columns separated by a tab ('\t'). Each word has been put on a separate line and there is an empty line after each sentences i.e the CoNLL format. The first item on each line is a word, the second is the named entity tag. The named entity tags have the format I-TYPE which means that the word is inside a phrase of type TYPE. For every multi-word expression like 'New York', the first word gets a tag B-TYPE and the subsequent words have tags I-TYPE, a word with tag O is not part of a phrase. The dataset is in the BIO tagging scheme. For more details, see
A collection of news article headlines in Hausa from VOA Hausa. Each headline is labeled with one of the following classes: Nigeria, Africa, World, Health or Politics. The dataset was presented in the paper: Hedderich, Adelani, Zhu, Alabi, Markus, Klakow: Transfer Learning and Distant Supervision for Multilingual Transformer Models: A Study on African Languages (EMNLP 2020).
HEAD-QA is a multi-choice HEAlthcare Dataset. The questions come from exams to access a specialized position in the Spanish healthcare system, and are challenging even for highly specialized humans. They are designed by the Ministerio de Sanidad, Consumo y Bienestar Social. The dataset contains questions about the following topics: medicine, nursing, psychology, chemistry, pharmacology and biology.
PUBHEALTH is a comprehensive dataset for explainable automated fact-checking of public health claims. Each instance in the PUBHEALTH dataset has an associated veracity label (true, false, unproven, mixture). Furthermore each instance in the dataset has an explanation text field. The explanation is a justification for which the claim has been assigned a particular veracity label. The dataset was created to explore fact-checking of difficult to verify claims i.e., those which require expertise from outside of the journalistics domain, in this case biomedical and public health expertise. It was also created in response to the lack of fact-checking datasets which provide gold standard natural language explanations for verdicts/labels. NOTE: There are missing labels in the dataset and we have replaced them with -1.
This repository contains a dump of thousands of public domain works in Hebrew, from Project Ben-Yehuda, in plaintext UTF-8 files, with and without diacritics (nikkud). The metadata (pseudocatalogue.csv) file is a list of titles, authors, genres, and file paths, to help you process the dump. All these works are in the public domain, so you are free to make any use of them, and do not need to ask for permission. There are 10078 files, 3181136 lines
HebrewSentiment is a data set consists of 12,804 user comments to posts on the official Facebook page of Israel’s president, Mr. Reuven Rivlin. In October 2015, we used the open software application Netvizz (Rieder, 2013) to scrape all the comments to all of the president’s posts in the period of June – August 2014, the first three months of Rivlin’s presidency.2 While the president’s posts aimed at reconciling tensions and called for tolerance and empathy, the sentiment expressed in the comments to the president’s posts was polarized between citizens who warmly thanked the president, and citizens that fiercely critiqued his policy. Of the 12,804 comments, 370 are neutral; 8,512 are positive, 3,922 negative. Data Annotation: A trained researcher examined each comment and determined its sentiment value, where comments with an overall positive sentiment were assigned the value 1, comments with an overall negative sentiment were assigned the value -1, and comments that are off-topic to the post’s content were assigned the value 0. We validated the coding scheme by asking a second trained researcher to code the same data. There was substantial agreement between raters (N of agreements: 10623, N of disagreements: 2105, Coehn’s Kappa = 0.697, p = 0).
HindEnCorp parallel texts (sentence-aligned) come from the following sources: Tides, which contains 50K sentence pairs taken mainly from news articles. This dataset was originally col- lected for the DARPA-TIDES surprise-language con- test in 2002, later refined at IIIT Hyderabad and provided for the NLP Tools Contest at ICON 2008 (Venkatapathy, 2008). Commentaries by Daniel Pipes contain 322 articles in English written by a journalist Daniel Pipes and translated into Hindi. EMILLE. This corpus (Baker et al., 2002) consists of three components: monolingual, parallel and annotated corpora. There are fourteen monolingual sub- corpora, including both written and (for some lan- guages) spoken data for fourteen South Asian lan- guages. The EMILLE monolingual corpora contain in total 92,799,000 words (including 2,627,000 words of transcribed spoken data for Bengali, Gujarati, Hindi, Punjabi and Urdu). The parallel corpus consists of 200,000 words of text in English and its accompanying translations into Hindi and other languages. Smaller datasets as collected by Bojar et al. (2010) include the corpus used at ACL 2005 (a subcorpus of EMILLE), a corpus of named entities from Wikipedia (crawled in 2009), and Agriculture domain parallel corpus.  For the current release, we are extending the parallel corpus using these sources: Intercorp (Čermák and Rosen,2012) is a large multilingual parallel corpus of 32 languages including Hindi. The central language used for alignment is Czech. Intercorp’s core texts amount to 202 million words. These core texts are most suitable for us because their sentence alignment is manually checked and therefore very reliable. They cover predominately short sto- ries and novels. There are seven Hindi texts in Inter- corp. Unfortunately, only for three of them the English translation is available; the other four are aligned only with Czech texts. The Hindi subcorpus of Intercorp contains 118,000 words in Hindi. TED talks 3 held in various languages, primarily English, are equipped with transcripts and these are translated into 102 languages. There are 179 talks for which Hindi translation is available. The Indic multi-parallel corpus (Birch et al., 2011; Post et al., 2012) is a corpus of texts from Wikipedia translated from the respective Indian language into English by non-expert translators hired over Mechanical Turk. The quality is thus somewhat mixed in many respects starting from typesetting and punctuation over capi- talization, spelling, word choice to sentence structure. A little bit of control could be in principle obtained from the fact that every input sentence was translated 4 times. We used the 2012 release of the corpus. is a software collaboration platform that hosts many open-source projects and facilitates also collaborative localization of the tools. We downloaded all revisions of all the hosted projects and extracted the localization (.po) files. Other smaller datasets. This time, we added Wikipedia entities as crawled in 2013 (including any morphological variants of the named entitity that appears on the Hindi variant of the Wikipedia page) and words, word examples and quotes from the Shabdkosh online dictionary.
The Hindi Discourse Analysis dataset is a corpus for analyzing discourse modes present in its sentences. It contains sentences from stories written by 11 famous authors from the 20th Century. 4-5 stories by each author have been selected which were available in the public domain resulting in a collection of 53 stories. Most of these short stories were originally written in Hindi but some of them were written in other Indian languages and later translated to Hindi.
To examine the cognitive processes of remembering and imagining and their traces in language, we introduce Hippocorpus, a dataset of 6,854 English diary-like short stories about recalled and imagined events. Using a crowdsourcing framework, we first collect recalled stories and summaries from workers, then provide these summaries to other workers who write imagined stories. Finally, months later, we collect a retold version of the recalled stories from a subset of recalled authors. Our dataset comes paired with author demographics (age, gender, race), their openness to experience, as well as some variables regarding the author's relationship to the event (e.g., how personal the event is, how often they tell its story, etc.).
The Hong Kong Cantonese Corpus (HKCanCor) comprise transcribed conversations recorded between March 1997 and August 1998. It contains recordings of spontaneous speech (51 texts) and radio programmes (42 texts), which involve 2 to 4 speakers, with 1 text of monologue. In total, the corpus contains around 230,000 Chinese words. The text is word-segmented, annotated with part-of-speech (POS) tags and romanised Cantonese pronunciation. Romanisation scheme - Linguistic Society of Hong Kong (LSHK) POS scheme - Peita-Fujitsu-Renmin Ribao (PRF) corpus (Duan et al., 2000), with extended tags for Cantonese-specific phenomena added by Luke and Wang (see original paper for details).
A Hope Speech dataset for Equality, Diversity and Inclusion (HopeEDI) containing user-generated comments from the social media platform YouTube with 28,451, 20,198 and 10,705 comments in English, Tamil and Malayalam, respectively, manually labelled as containing hope speech or not.
HoVer is an open-domain, many-hop fact extraction and claim verification dataset built upon the Wikipedia corpus. The original 2-hop claims are adapted from question-answer pairs from HotpotQA. It is collected by a team of NLP researchers at UNC Chapel Hill and Verisk Analytics.
The hrenWaC corpus version 2.0 consists of parallel Croatian-English texts crawled from the .hr top-level domain for Croatia. The corpus was built with Spidextor (, a tool that glues together the output of SpiderLing used for crawling and Bitextor used for bitext extraction. The accuracy of the extracted bitext on the segment level is around 80% and on the word level around 84%.
The Croatian web corpus hrWaC was built by crawling the .hr top-level domain in 2011 and again in 2014. The corpus was near-deduplicated on paragraph level, normalised via diacritic restoration, morphosyntactically annotated and lemmatised. The corpus is shuffled by paragraphs. Each paragraph contains metadata on the URL, domain and language identification (Croatian vs. Serbian). Version 2.0 of this corpus is described in Version 2.1 contains newer and better linguistic annotations.
Existing question answering datasets focus on dealing with homogeneous information, based either only on text or KB/Table information alone. However, as human knowledge is distributed over heterogeneous forms, using homogeneous information alone might lead to severe coverage problems. To fill in the gap, we present HybridQA, a new large-scale question-answering dataset that requires reasoning on heterogeneous information. Each question is aligned with a Wikipedia table and multiple free-form corpora linked with the entities in the table. The questions are designed to aggregate both tabular information and text information, i.e., lack of either form would render the question unanswerable.
The CLICK-ID dataset is a collection of Indonesian news headlines that was collected from 12 local online news publishers; detikNews, Fimela, Kapanlagi, Kompas, Liputan6, Okezone, Posmetro-Medan, Republika, Sindonews, Tempo, Tribunnews, and Wowkeren. This dataset is comprised of mainly two parts; (i) 46,119 raw article data, and (ii) 15,000 clickbait annotated sample headlines. Annotation was conducted with 3 annotator examining each headline. Judgment were based only on the headline. The majority then is considered as the ground truth. In the annotated sample, our annotation shows 6,290 clickbait and 8,710 non-clickbait.
In this paper, we introduce a large-scale Indonesian summarization dataset. We harvest articles from this http URL, an online news portal, and obtain 215,827 document-summary pairs. We leverage pre-trained language models to develop benchmark extractive and abstractive summarization methods over the dataset with multilingual and monolingual BERT-based models. We include a thorough error analysis by examining machine-generated summaries that have low ROUGE scores, and expose both issues with ROUGE it-self, as well as with extractive and abstractive summarization models.
Nergrit Corpus is a dataset collection for Indonesian Named Entity Recognition, Statement Extraction, and Sentiment Analysis. id_nergrit_corpus is the Named Entity Recognition of this dataset collection which contains 18 entities as follow: 'CRD': Cardinal 'DAT': Date 'EVT': Event 'FAC': Facility 'GPE': Geopolitical Entity 'LAW': Law Entity (such as Undang-Undang) 'LOC': Location 'MON': Money 'NOR': Political Organization 'ORD': Ordinal 'ORG': Organization 'PER': Person 'PRC': Percent 'PRD': Product 'QTY': Quantity 'REG': Religion 'TIM': Time 'WOA': Work of Art 'LAN': Language
The dataset contains around 500K articles (136M of words) from 7 Indonesian newspapers: Detik, Kompas, Tempo, CNN Indonesia, Sindo, Republika and Poskota. The articles are dated between 1st January 2018 and 20th August 2018 (with few exceptions dated earlier). The size of uncompressed 500K json files (newspapers-json.tgz) is around 2.2GB, and the cleaned uncompressed in a big text file (newspapers.txt.gz) is about 1GB. The original source in Google Drive contains also a dataset in html format which include raw data (pictures, css, javascript, ...) from the online news website
Parallel Text Corpora for Multi-Domain Translation System created by BPPT (Indonesian Agency for the Assessment and Application of Technology) for PAN Localization Project (A Regional Initiative to Develop Local Language Computing Capacity in Asia). The dataset contains around 24K sentences divided in 4 difference topics (Economic, international, Science and Technology and Sport).
Large Movie translated Urdu Reviews Dataset. This is a dataset for binary sentiment classification containing substantially more data than previous benchmark datasets. We provide a set of 40,000 highly polar movie reviews for training, and 10,000 for testing. To increase the availability of sentiment analysis dataset for a low recourse language like Urdu, we opted to use the already available IMDB Dataset. we have translated this dataset using google translator. This is a binary classification dataset having two classes as positive and negative. The reason behind using this dataset is high polarity for each class. It contains 50k samples equally divided in two classes.
Over >25k semiautomatically generated sentence pairs illustrating well-studied pragmatic inference types. IMPPRES is an NLI dataset following the format of SNLI (Bowman et al., 2015), MultiNLI (Williams et al., 2018) and XNLI (Conneau et al., 2018), which was created to evaluate how well trained NLI models recognize several classes of presuppositions and scalar implicatures.
A dataset of about 20k questions that are elicited from readers as they naturally read through a document sentence by sentence. Compared to existing datasets, INQUISITIVE questions target more towards high-level (semantic and discourse) comprehension of text. Because these questions are generated while the readers are processing the information, the questions directly communicate gaps between the reader’s and writer’s knowledge about the events described in the text, and are not necessarily answered in the document itself. This type of question reflects a real-world scenario: if one has questions during reading, some of them are answered by the text later on, the rest are not, but any of them would help further the reader’s understanding at the particular point when they asked it. This resource could enable question generation models to simulate human-like curiosity and cognitive processing, which may open up a new realm of applications.
It is a Turkish news data set consisting of 273601 news in 17 categories, compiled from print media and news websites between 2010 and 2017 by the Interpress ( media monitoring company.
The Kannada news dataset contains only the headlines of news article in three categories: Entertainment, Tech, and Sports. The data set contains around 6300 news article headlines which collected from Kannada news websites. The data set has been cleaned and contains train and test set using which can be used to benchmark classification models in Kannada.
KdConv is a Chinese multi-domain Knowledge-driven Conversionsation dataset, grounding the topics in multi-turn conversations to knowledge graphs. KdConv contains 4.5K conversations from three domains (film, music, and travel), and 86K utterances with an average turn number of 19.0. These conversations contain in-depth discussions on related topics and natural transition between multiple topics, while the corpus can also used for exploration of transfer learning and domain adaptation.\
A parallel corpus of KDE4 localization files (v.2). 92 languages, 4,099 bitexts total number of files: 75,535 total number of tokens: 60.75M total number of sentence fragments: 8.89M
Data-To-Text Generation involves converting knowledge graph (KG) triples of the form (subject, relation, object) into a natural language sentence(s). This dataset consists of English KG data converted into paired natural language text. The generated corpus consists of ∼18M sentences spanning ∼45M triples with ∼1500 distinct relations.
KILT tasks training and evaluation data. - [FEVER]( | Fact Checking | fever - [AIDA CoNLL-YAGO]( | Entity Linking | aidayago2 - [WNED-WIKI]( | Entity Linking | wned - [WNED-CWEB]( | Entity Linking | cweb - [T-REx]( | Slot Filling | trex - [Zero-Shot RE]( | Slot Filling | structured_zeroshot - [Natural Questions]( | Open Domain QA | nq - [HotpotQA]( | Open Domain QA | hotpotqa - [TriviaQA]( | Open Domain QA | triviaqa - [ELI5]( | Open Domain QA | eli5 - [Wizard of Wikipedia]( | Dialogue | wow To finish linking TriviaQA questions to the IDs provided, follow the instructions [here](
This dataset is designed to identify speaker intention based on real-life spoken utterance in Korean into one of 7 categories: fragment, description, question, command, rhetorical question, rhetorical command, utterances.
This is a Korean paired question dataset containing labels indicating whether two questions in a given pair are semantically identical. This dataset was used to evaluate the performance of [KoGPT2]( on a phrase detection downstream task.
This new dataset is designed to extract intent from non-canonical directives which will help dialog managers extract intent from user dialog that may have no clear objective or are paraphrased forms of utterances.
This dataset contains over 63,000 book reviews in Arabic.It is the largest sentiment analysis dataset for Arabic to-date.The book reviews were harvested from the website Goodreads during the month or March 2013.Each book review comes with the goodreads review id, the user id, the book id, the rating (1 to 5) and the text of the review.
The Large Spanish Corpus is a compilation of 15 unlabelled Spanish corpora spanning Wikipedia to European parliament notes. Each config contains the data corresponding to a different corpus. For example, "all_wiki" only includes examples from Spanish Wikipedia. By default, the config is set to "combined" which loads all the corpora; with this setting you can also specify the number of samples to return per corpus by configuring the "split" argument.
LeNER-Br is a Portuguese language dataset for named entity recognition applied to legal documents. LeNER-Br consists entirely of manually annotated legislation and legal cases texts and contains tags for persons, locations, time entities, organizations, legislation and legal cases. To compose the dataset, 66 legal documents from several Brazilian Courts were collected. Courts of superior and state levels were considered, such as Supremo Tribunal Federal, Superior Tribunal de Justiça, Tribunal de Justiça de Minas Gerais and Tribunal de Contas da União. In addition, four legislation documents were collected, such as "Lei Maria da Penha", giving a total of 70 documents
LIAR is a dataset for fake news detection with 12.8K human labeled short statements from's API, and each statement is evaluated by a editor for its truthfulness. The distribution of labels in the LIAR dataset is relatively well-balanced: except for 1,050 pants-fire cases, the instances for all other labels range from 2,063 to 2,638. In each case, the labeler provides a lengthy analysis report to ground each judgment.
Motion recognition is one of the basic cognitive capabilities of many life forms, yet identifying motion of physical entities in natural language have not been explored extensively and empirically. Literal-Motion-in-Text (LiMiT) dataset, is a large human-annotated collection of English text sentences describing physical occurrence of motion, with annotated physical entities in motion.
This is LiveQA, a Chinese dataset constructed from play-by-play live broadcast. It contains 117k multiple-choice questions written by human commentators for over 1,670 NBA games, which are collected from the Chinese Hupu website.
LST20 Corpus is a dataset for Thai language processing developed by National Electronics and Computer Technology Center (NECTEC), Thailand. It offers five layers of linguistic annotation: word boundaries, POS tagging, named entities, clause boundaries, and sentence boundaries. At a large scale, it consists of 3,164,002 words, 288,020 named entities, 248,181 clauses, and 74,180 sentences, while it is annotated with 16 distinct POS tags. All 3,745 documents are also annotated with one of 15 news genres. Regarding its sheer size, this dataset is considered large enough for developing joint neural models for NLP. Manually download at
Mac-Morpho is a corpus of Brazilian Portuguese texts annotated with part-of-speech tags. Its first version was released in 2003 [1], and since then, two revisions have been made in order to improve the quality of the resource [2, 3]. The corpus is available for download split into train, development and test sections. These are 76%, 4% and 20% of the corpus total, respectively (the reason for the unusual numbers is that the corpus was first split into 80%/20% train/test, and then 5% of the train section was set aside for development). This split was used in [3], and new POS tagging research with Mac-Morpho is encouraged to follow it in order to make consistent comparisons possible. [1] Aluísio, S., Pelizzoni, J., Marchi, A.R., de Oliveira, L., Manenti, R., Marquiafável, V. 2003. An account of the challenge of tagging a reference corpus for brazilian portuguese. In: Proceedings of the 6th International Conference on Computational Processing of the Portuguese Language. PROPOR 2003 [2] Fonseca, E.R., Rosa, J.L.G. 2013. Mac-morpho revisited: Towards robust part-of-speech. In: Proceedings of the 9th Brazilian Symposium in Information and Human Language Technology – STIL [3] Fonseca, E.R., Aluísio, Sandra Maria, Rosa, J.L.G. 2015. Evaluating word embeddings and a revised corpus for part-of-speech tagging in Portuguese. Journal of the Brazilian Computer Society.
MC-TACO (Multiple Choice TemporAl COmmonsense) is a dataset of 13k question-answer pairs that require temporal commonsense comprehension. A system receives a sentence providing context information, a question designed to require temporal commonsense knowledge, and multiple candidate answers. More than one candidate answer can be plausible. The task is framed as binary classification: givent he context, the question, and the candidate answer, the task is to determine whether the candidate answer is plausible ("yes") or not ("no").
Machine learning models are trained to find patterns in data. NLP models can inadvertently learn socially undesirable patterns when training on gender biased text. In this work, we propose a general framework that decomposes gender bias in text along several pragmatic and semantic dimensions: bias from the gender of the person being spoken about, bias from the gender of the person being spoken to, and bias from the gender of the speaker. Using this fine-grained framework, we automatically annotate eight large scale datasets with gender information. In addition, we collect a novel, crowdsourced evaluation benchmark of utterance-level gender rewrites. Distinguishing between gender bias along multiple dimensions is important, as it enables us to train finer-grained gender bias classifiers. We show our classifiers prove valuable for a variety of important applications, such as controlling for gender bias in generative models, detecting gender bias in arbitrary text, and shed light on offensive language in terms of genderedness.
MedHop is based on research paper abstracts from PubMed, and the queries are about interactions between pairs of drugs. The correct answer has to be inferred by combining information from a chain of reactions of drugs and proteins.
A large medical text dataset (14Go) curated to 4Go for abbreviation disambiguation, designed for natural language understanding pre-training in the medical domain. For example, DHF can be disambiguated to dihydrofolate, diastolic heart failure, dengue hemorragic fever or dihydroxyfumarate
The MedDialog dataset (English) contains conversations (in English) between doctors and patients.It has 0.26 million dialogues. The data is continuously growing and more dialogues will be added. The raw dialogues are from and All copyrights of the data belong to and
MENYO-20k is a multi-domain parallel dataset with texts obtained from news articles, ted talks, movie transcripts, radio transcripts, science and technology texts, and other short articles curated from the web and professional translators. The dataset has 20,100 parallel sentences split into 10,070 training sentences, 3,397 development sentences, and 6,633 test sentences (3,419 multi-domain, 1,714 news domain, and 1,500 ted talks speech transcript domain). The development and test sets are available upon request.
MetaLWOz: A Dataset of Multi-Domain Dialogues for the Fast Adaptation of Conversation Models. We introduce the Meta-Learning Wizard of Oz (MetaLWOz) dialogue dataset for developing fast adaptation methods for conversation models. This data can be used to train task-oriented dialogue models, specifically to develop methods to quickly simulate user responses with a small amount of data. Such fast-adaptation models fall into the research areas of transfer learning and meta learning. The dataset consists of 37,884 crowdsourced dialogues recorded between two human users in a Wizard of Oz setup, in which one was instructed to behave like a bot, and the other a true human user. The users are assigned a task belonging to a particular domain, for example booking a reservation at a particular restaurant, and work together to complete the task. Our dataset spans 47 domains having 227 tasks total. Dialogues are a minimum of 10 turns long.
The dataset consists of tweets belonging to #MeToo movement on Twitter, labelled into different categories. Due to Twitter's development policies, we only provide the tweet ID's and corresponding labels, other data can be fetched via Twitter API. The data has been labelled by experts, with the majority taken into the account for deciding the final label. We provide these labels for each of the tweets. The labels provided for each data point includes -- Relevance, Directed Hate, Generalized Hate, Sarcasm, Allegation, Justification, Refutation, Support, Oppose
Arabic Poetry Metric Classification. 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.The train dataset contains 47,124 records and the test dataset contains 8316 records.
We introduce MKQA, an open-domain question answering evaluation set comprising 10k question-answer pairs sampled from the Google Natural Questions dataset, aligned across 26 typologically diverse languages (260k question-answer pairs in total). For each query we collected new passage-independent answers. These queries and answers were then human translated into 25 Non-English languages.
Posing reading comprehension as a generation problem provides a great deal of flexibility, allowing for open-ended questions with few restrictions on possible answers. However, progress is impeded by existing generation metrics, which rely on token overlap and are agnostic to the nuances of reading comprehension. To address this, we introduce a benchmark for training and evaluating generative reading comprehension metrics: MOdeling Correctness with Human Annotations. MOCHA contains 40K human judgement scores on model outputs from 6 diverse question answering datasets and an additional set of minimal pairs for evaluation. Using MOCHA, we train an evaluation metric: LERC, a Learned Evaluation metric for Reading Comprehension, to mimic human judgement scores.
The Microsoft Terminology Collection can be used to develop localized versions of applications that integrate with Microsoft products. It can also be used to integrate Microsoft terminology into other terminology collections or serve as a base IT glossary for language development in the nearly 100 languages available. Terminology is provided in .tbx format, an industry standard for terminology exchange.
The database is derived from the NCI PID Pathway Interaction Database, and the textual mentions are extracted from cooccurring pairs of genes in PubMed abstracts, processed and annotated by Literome (Poon et al. 2014). This dataset was used in the paper “Compositional Learning of Embeddings for Relation Paths in Knowledge Bases and Text” (Toutanova, Lin, Yih, Poon, and Quirk, 2016).
Recent work in semantic parsing for question answering has focused on long and complicated questions, many of which would seem unnatural if asked in a normal conversation between two humans. In an effort to explore a conversational QA setting, we present a more realistic task: answering sequences of simple but inter-related questions. We created SQA by asking crowdsourced workers to decompose 2,022 questions from WikiTableQuestions (WTQ), which contains highly-compositional questions about tables from Wikipedia. We had three workers decompose each WTQ question, resulting in a dataset of 6,066 sequences that contain 17,553 questions in total. Each question is also associated with answers in the form of cell locations in the tables.
The Third International Chinese Language Processing Bakeoff was held in Spring 2006 to assess the state of the art in two important tasks: word segmentation and named entity recognition. Twenty-nine groups submitted result sets in the two tasks across two tracks and a total of five corpora. We found strong results in both tasks as well as continuing challenges. MSRA NER is one of the provided dataset. There are three types of NE, PER (person), ORG (organization) and LOC (location). The dataset is in the BIO scheme. For more details see
MultiBooked is a corpus of Basque and Catalan Hotel Reviews Annotated for Aspect-level Sentiment Classification. The corpora are compiled from hotel reviews taken mainly from The corpora are in Kaf/Naf format, which is an xml-style stand-off format that allows for multiple layers of annotation. Each review was sentence- and word-tokenized and lemmatized using Freeling for Catalan and ixa-pipes for Basque. Finally, for each language two annotators annotated opinion holders, opinion targets, and opinion expressions for each review, following the guidelines set out in the OpeNER project.
Parallel corpora from Web Crawls collected in the ParaCrawl project and further processed for making it a multi-parallel corpus by pivoting via English. Here we only provide the additional language pairs that came out of pivoting. The bitexts for English are available from the ParaCrawl release. 40 languages, 669 bitexts total number of files: 40 total number of tokens: 10.14G total number of sentence fragments: 505.48M Please, acknowledge the ParaCrawl project at This version is derived from the original release at their website adjusted for redistribution via the OPUS corpus collection. Please, acknowledge OPUS as well for this service.
Multi-Domain Wizard-of-Oz dataset (MultiWOZ), a fully-labeled collection of human-human written conversations spanning over multiple domains and topics. MultiWOZ 2.1 (Eric et al., 2019) identified and fixed many erroneous annotations and user utterances in the original version, resulting in an improved version of the dataset. MultiWOZ 2.2 is a yet another improved version of this dataset, which identifies and fizes dialogue state annotation errors across 17.3% of the utterances on top of MultiWOZ 2.1 and redefines the ontology by disallowing vocabularies of slots with a large number of possible values (e.g., restaurant name, time of booking) and introducing standardized slot span annotations for these slots.
Multi-XScience, a large-scale multi-document summarization dataset created from scientific articles. Multi-XScience introduces a challenging multi-document summarization task: writing the related-work section of a paper based on its abstract and the articles it references.
Our goal is to build systems that collaborate with people by exchanging information through natural language and reasoning over structured knowledge base. In the MutualFriend task, two agents, A and B, each have a private knowledge base, which contains a list of friends with multiple attributes (e.g., name, school, major, etc.). The agents must chat with each other to find their unique mutual friend.
The Myanmar news dataset contains article snippets in four categories: Business, Entertainment, Politics, and Sport. These were collected in October 2017 by Aye Hninn Khine
The NarrativeQA dataset for question answering on long documents (movie scripts, books). It includes the list of documents with Wikipedia summaries, links to full stories, and questions and answers.
This paper presents the disease name and concept annotations of the NCBI disease corpus, a collection of 793 PubMed abstracts fully annotated at the mention and concept level to serve as a research resource for the biomedical natural language processing community. Each PubMed abstract was manually annotated by two annotators with disease mentions and their corresponding concepts in Medical Subject Headings (MeSH®) or Online Mendelian Inheritance in Man (OMIM®). Manual curation was performed using PubTator, which allowed the use of pre-annotations as a pre-step to manual annotations. Fourteen annotators were randomly paired and differing annotations were discussed for reaching a consensus in two annotation phases. In this setting, a high inter-annotator agreement was observed. Finally, all results were checked against annotations of the rest of the corpus to assure corpus-wide consistency. For more details, see: The original dataset can be downloaded from: This dataset has been converted to CoNLL format for NER using the following tool: Note: there is a duplicate document (PMID 8528200) in the original data, and the duplicate is recreated in the converted data.
The development of linguistic resources for use in natural language processingis of utmost importance for the continued growth of research anddevelopment in the field, especially for resource-scarce languages. In this paper we describe the process and challenges of simultaneouslydevelopingmultiple linguistic resources for ten of the official languages of South Africa. The project focussed on establishing a set of foundational resources that can foster further development of both resources and technologies for the NLP industry in South Africa. The development efforts during the project included creating monolingual unannotated corpora, of which a subset of the corpora for each language was annotated on token, orthographic, morphological and morphosyntactic layers. The annotated subsetsincludes both development and test setsand were used in the creation of five core-technologies, viz. atokeniser, sentenciser,lemmatiser, part of speech tagger and morphological decomposer for each language. We report on the quality of these tools for each language and provide some more context of the importance of the resources within the South African context.
This dataset provides version 1115 of the belief extracted by CMU's Never Ending Language Learner (NELL) and version 1110 of the candidate belief extracted by NELL. See NELL is an open information extraction system that attempts to read the Clueweb09 of 500 million web pages ( and general web searches. The dataset has 4 configurations: nell_belief, nell_candidate, nell_belief_sentences, and nell_candidate_sentences. nell_belief is certainties of belief are lower. The two sentences config extracts the CPL sentence patterns filled with the applicable 'best' literal string for the entities filled into the sentence patterns. And also provides sentences found using web searches containing the entities and relationships. There are roughly 21M entries for nell_belief_sentences, and 100M sentences for nell_candidate_sentences.
Neural-Code-Search-Evaluation-Dataset presents an evaluation dataset consisting of natural language query and code snippet pairs and a search corpus consisting of code snippets collected from the most popular Android repositories on GitHub.
A parallel corpus of News Commentaries provided by WMT for training SMT. The source is taken from CASMACAT: 12 languages, 63 bitexts total number of files: 61,928 total number of tokens: 49.66M total number of sentence fragments: 1.93M
First benchmark dataset for sentence entailment in the low-resource Filipino language. Constructed through exploting the structure of news articles. Contains 600,000 premise-hypothesis pairs, in 70-15-15 split for training, validation, and testing.
NewsQA is a challenging machine comprehension dataset of over 100,000 human-generated question-answer pairs. Crowdworkers supply questions and answers based on a set of over 10,000 news articles from CNN, with answers consisting of spans of text from the corresponding articles.
The NKJP-NER is based on a human-annotated part of National Corpus of Polish (NKJP). We extracted sentences with named entities of exactly one type. The task is to predict the type of the named entity.
Named entities Recognition dataset for Norwegian. It is a version of the Universal Dependency (UD) Treebank for both Bokmål and Nynorsk (UDN) where all proper nouns have been tagged with their type according to the NER tagging scheme. UDN is a converted version of the Norwegian Dependency Treebank into the UD scheme.
This is a movie review dataset in the Korean language. Reviews were scraped from Naver movies. The dataset construction is based on the method noted in Large movie review dataset from Maas et al., 2011.
Fused Head constructions are noun phrases in which the head noun is missing and is said to be "fused" with its dependent modifier. This missing information is implicit and is important for sentence understanding.The missing heads are easily filled in by humans, but pose a challenge for computational models. For example, in the sentence: "I bought 5 apples but got only 4.", 4 is a Fused-Head, and the missing head is apples, which appear earlier in the sentence. This is a crowd-sourced dataset of 10k numerical fused head examples (1M tokens).
The researchers of OCLAR Marwan et al. (2019), they gathered Arabic costumer reviews from Google reviewsa and Zomato website ( on wide scope of domain, including restaurants, hotels, hospitals, local shops, etc.The corpus finally contains 3916 reviews in 5-rating scale. For this research purpose, the positive class considers rating stars from 5 to 3 of 3465 reviews, and the negative class is represented from values of 1 and 2 of about 451 texts.