The viewer is disabled because this dataset repo requires arbitrary Python code execution. Please consider removing the loading script and relying on automated data support. If this is not possible, please open a discussion for direct help.

Leipzig Corpora Collection

The Leipzig Corpora Collection presents corpora in different languages using the same format and comparable sources. All data are available as plain text files and can be imported into a MySQL database by using the provided import script. They are intended both for scientific use by corpus linguists as well as for applications such as knowledge extraction programs. The corpora are identical in format and similar in size and content. They contain randomly selected sentences in the language of the corpus and are available in sizes from 10,000 sentences up to 1 million sentences. The sources are either newspaper texts or texts randomly collected from the web. The texts are split into sentences. Non-sentences and foreign language material was removed. Because word co-occurrence information is useful for many applications, these data are precomputed and included as well. For each word, the most significant words appearing as immediate left or right neighbor or appearing anywhere within the same sentence are given. More information about the format and content of these files can be found here. The corpora are automatically collected from carefully selected public sources without considering in detail the content of the contained text. No responsibility is taken for the content of the data. In particular, the views and opinions expressed in specific parts of the data remain exclusively with the authors.

Dataset Usage

Links

A "links" subset contains URLs with corresponding language and id (based on https://corpora.uni-leipzig.de/)

from datasets import load_dataset

ds = load_dataset("imvladikon/leipzig_corpora_collection", "links")
for row in ds["train"]:
    print(row)
{'id': '0', 'data_id': '0', 'url': 'https://downloads.wortschatz-leipzig.de/corpora/ara_news_2005-2009_10K.tar.gz', 'language': 'Arabic', 'language_short': 'ara', 'year': '2005', 'size': '10K'}
{'id': '1', 'data_id': '1', 'url': 'https://downloads.wortschatz-leipzig.de/corpora/ara_news_2005-2009_30K.tar.gz', 'language': 'Arabic', 'language_short': 'ara', 'year': '2005', 'size': '30K'}
{'id': '2', 'data_id': '2', 'url': 'https://downloads.wortschatz-leipzig.de/corpora/ara_news_2005-2009_100K.tar.gz', 'language': 'Arabic', 'language_short': 'ara', 'year': '2005', 'size': '100K'}
....

where is possible to choose specific data_id to load a specific dataset, where data_id is name of the subset

Links possible to filter according to metdata attributes:

links = load_dataset("imvladikon/leipzig_corpora_collection", "links", split="train")

english_2019 = links.filter(lambda x: x["language"] == "English" and x["year"] == "2019")

for sample in english_2019:
    print(sample)
{'id': '277', 'data_id': 'eng_news_2019_10K', 'url': 'https://downloads.wortschatz-leipzig.de/corpora/eng_news_2019_10K.tar.gz', 'language': 'English', 'language_short': 'eng', 'year': '2019', 'size': '10K'}
{'id': '278', 'data_id': 'eng_news_2019_30K', 'url': 'https://downloads.wortschatz-leipzig.de/corpora/eng_news_2019_30K.tar.gz', 'language': 'English', 'language_short': 'eng', 'year': '2019', 'size': '30K'}
{'id': '279', 'data_id': 'eng_news_2019_100K', 'url': 'https://downloads.wortschatz-leipzig.de/corpora/eng_news_2019_100K.tar.gz', 'language': 'English', 'language_short': 'eng', 'year': '2019', 'size': '100K'}
{'id': '280', 'data_id': 'eng_news_2019_300K', 'url': 'https://downloads.wortschatz-leipzig.de/corpora/eng_news_2019_300K.tar.gz', 'language': 'English', 'language_short': 'eng', 'year': '2019', 'size': '300K'}
{'id': '281', 'data_id': 'eng_news_2019_1M', 'url': 'https://downloads.wortschatz-leipzig.de/corpora/eng_news_2019_1M.tar.gz', 'language': 'English', 'language_short': 'eng', 'year': '2019', 'size': '1M'}
{'id': '541', 'data_id': 'eng-za_web_2019_10K', 'url': 'https://downloads.wortschatz-leipzig.de/corpora/eng-za_web_2019_10K.tar.gz', 'language': 'English', 'language_short': 'eng', 'year': '2019', 'size': '10K'}
{'id': '542', 'data_id': 'eng-za_web_2019_30K', 'url': 'https://downloads.wortschatz-leipzig.de/corpora/eng-za_web_2019_30K.tar.gz', 'language': 'English', 'language_short': 'eng', 'year': '2019', 'size': '30K'}
{'id': '543', 'data_id': 'eng-za_web_2019_100K', 'url': 'https://downloads.wortschatz-leipzig.de/corpora/eng-za_web_2019_100K.tar.gz', 'language': 'English', 'language_short': 'eng', 'year': '2019', 'size': '100K'}
{'id': '544', 'data_id': 'eng-za_web_2019_300K', 'url': 'https://downloads.wortschatz-leipzig.de/corpora/eng-za_web_2019_300K.tar.gz', 'language': 'English', 'language_short': 'eng', 'year': '2019', 'size': '300K'}
{'id': '545', 'data_id': 'eng-za_web_2019_1M', 'url': 'https://downloads.wortschatz-leipzig.de/corpora/eng-za_web_2019_1M.tar.gz', 'language': 'English', 'language_short': 'eng', 'year': '2019', 'size': '1M'}

Corpus

after selecting data_id, let's say heb_wikipedia_2021_1M, we could load it:

dataset_he = load_dataset("imvladikon/leipzig_corpora_collection", "heb_wikipedia_2021_1M", split="train")
for row in dataset_he:
    print(row)

another example:

dataset_en = load_dataset("imvladikon/leipzig_corpora_collection", "eng-simple_wikipedia_2021_300K", split="train")
print(dataset_en[76576])

sample:

{'id': '79214', 'sentence': 'He was a member of the assembly from 1972 to 1977.'}

Citation

If you use one of these corpora in your work, please, to cite this work:

@inproceedings{goldhahn-etal-2012-building,
    title = "Building Large Monolingual Dictionaries at the {L}eipzig Corpora Collection: From 100 to 200 Languages",
    author = "Goldhahn, Dirk  and
      Eckart, Thomas  and
      Quasthoff, Uwe",
    editor = "Calzolari, Nicoletta  and
      Choukri, Khalid  and
      Declerck, Thierry  and
      Do{\u{g}}an, Mehmet U{\u{g}}ur  and
      Maegaard, Bente  and
      Mariani, Joseph  and
      Moreno, Asuncion  and
      Odijk, Jan  and
      Piperidis, Stelios",
    booktitle = "Proceedings of the Eighth International Conference on Language Resources and Evaluation ({LREC}'12)",
    month = may,
    year = "2012",
    address = "Istanbul, Turkey",
    publisher = "European Language Resources Association (ELRA)",
    url = "http://www.lrec-conf.org/proceedings/lrec2012/pdf/327_Paper.pdf",
    pages = "759--765",
    abstract = "The Leipzig Corpora Collection offers free online access to 136 monolingual dictionaries enriched with statistical information. In this paper we describe current advances of the project in collecting and processing text data automatically for a large number of languages. Our main interest lies in languages of “low density”, where only few text data exists online. The aim of this approach is to create monolingual dictionaries and statistical information for a high number of new languages and to expand the existing dictionaries, opening up new possibilities for linguistic typology and other research. Focus of this paper will be set on the infrastructure for the automatic acquisition of large amounts of monolingual text in many languages from various sources. Preliminary results of the collection of text data will be presented. The mainly language-independent framework for preprocessing, cleaning and creating the corpora and computing the necessary statistics will also be depicted.",
}
Downloads last month
612
Edit dataset card