Dataset Card for "lince"

Dataset Summary

LinCE is a centralized Linguistic Code-switching Evaluation benchmark (https://ritual.uh.edu/lince/) that contains data for training and evaluating NLP systems on code-switching tasks.

Supported Tasks and Leaderboards

More Information Needed

Languages

More Information Needed

Dataset Structure

Data Instances

lid_hineng

• Size of downloaded dataset files: 0.43 MB
• Size of the generated dataset: 2.39 MB
• Total amount of disk used: 2.82 MB

An example of 'validation' looks as follows.

{
    "idx": 0,
    "lid": ["other", "other", "lang1", "lang1", "lang1", "other", "lang1", "lang1", "lang1", "lang1", "lang1", "lang1", "lang1", "mixed", "lang1", "lang1", "other"],
    "words": ["@ZahirJ", "@BinyavangaW", "Loved", "the", "ending", "!", "I", "could", "have", "offered", "you", "some", "ironic", "chai-tea", "for", "it", ";)"]
}

lid_msaea

• Size of downloaded dataset files: 0.81 MB
• Size of the generated dataset: 4.89 MB
• Total amount of disk used: 5.69 MB

An example of 'train' looks as follows.

This example was too long and was cropped:

{
    "idx": 0,
    "lid": ["ne", "lang2", "other", "lang2", "lang2", "other", "other", "lang2", "lang2", "lang2", "lang2", "lang2", "lang2", "lang2", "lang2", "lang2", "lang2", "lang2", "lang2", "lang2", "lang2", "other", "lang2", "lang2", "lang2", "ne", "lang2", "lang2"],
    "words": "[\"علاء\", \"بخير\", \"،\", \"معنوياته\", \"كويسة\", \".\", \"..\", \"اسخف\", \"حاجة\", \"بس\", \"ان\", \"كل\", \"واحد\", \"منهم\", \"بييقى\", \"مقفول\", \"عليه\"..."
}

lid_nepeng

• Size of downloaded dataset files: 0.55 MB
• Size of the generated dataset: 3.21 MB
• Total amount of disk used: 3.75 MB

An example of 'validation' looks as follows.

{
    "idx": 1,
    "lid": ["other", "lang2", "lang2", "lang2", "lang2", "lang1", "lang1", "lang1", "lang1", "lang1", "lang2", "lang2", "other", "mixed", "lang2", "lang2", "other", "other", "other", "other"],
    "words": ["@nirvikdada", "la", "hamlai", "bhetna", "paayeko", "will", "be", "your", "greatest", "gift", "ni", "dada", ";P", "#TreatChaiyo", "j", "hos", ";)", "@zappylily", "@AsthaGhm", "@ayacs_asis"]
}

lid_spaeng

• Size of downloaded dataset files: 1.18 MB
• Size of the generated dataset: 6.83 MB
• Total amount of disk used: 8.01 MB

An example of 'train' looks as follows.

{
    "idx": 0,
    "lid": ["other", "other", "lang1", "lang1", "lang1", "other", "lang1", "lang1"],
    "words": ["11:11", ".....", "make", "a", "wish", ".......", "night", "night"]
}

ner_hineng

• Size of downloaded dataset files: 0.14 MB
• Size of the generated dataset: 0.79 MB
• Total amount of disk used: 0.92 MB

An example of 'train' looks as follows.

{
    "idx": 1,
    "lid": ["en", "en", "en", "en", "en", "en", "hi", "hi", "hi", "hi", "hi", "hi", "hi", "en", "en", "en", "en", "rest"],
    "ner": ["O", "O", "O", "O", "O", "O", "O", "O", "O", "O", "O", "B-PERSON", "I-PERSON", "O", "O", "O", "B-PERSON", "I-PERSON"],
    "words": ["I", "liked", "a", "@YouTube", "video", "https://t.co/DmVqhZbdaI", "Kabhi", "Palkon", "Pe", "Aasoon", "Hai-", "Kishore", "Kumar", "-Vocal", "Cover", "By", "Stephen", "Qadir"]
}

Data Fields

The data fields are the same among all splits.

lid_hineng

• idx: a int32 feature.
• words: a list of string features.
• lid: a list of string features.

lid_msaea

• idx: a int32 feature.
• words: a list of string features.
• lid: a list of string features.

lid_nepeng

• idx: a int32 feature.
• words: a list of string features.
• lid: a list of string features.

lid_spaeng

• idx: a int32 feature.
• words: a list of string features.
• lid: a list of string features.

ner_hineng

• idx: a int32 feature.
• words: a list of string features.
• lid: a list of string features.
• ner: a list of string features.

Data Splits

Dataset Creation

Curation Rationale

More Information Needed

Source Data

Initial Data Collection and Normalization

More Information Needed

Who are the source language producers?

More Information Needed

Annotations

Annotation process

More Information Needed

Who are the annotators?

More Information Needed

Personal and Sensitive Information

More Information Needed

Considerations for Using the Data

Social Impact of Dataset

More Information Needed

Discussion of Biases

More Information Needed

Other Known Limitations

More Information Needed

Additional Information

Dataset Curators

More Information Needed

Licensing Information

More Information Needed

Citation Information

@inproceedings{aguilar-etal-2020-lince,
    title = "{L}in{CE}: A Centralized Benchmark for Linguistic Code-switching Evaluation",
    author = "Aguilar, Gustavo  and
      Kar, Sudipta  and
      Solorio, Thamar",
    booktitle = "Proceedings of The 12th Language Resources and Evaluation Conference",
    month = may,
    year = "2020",
    address = "Marseille, France",
    publisher = "European Language Resources Association",
    url = "https://www.aclweb.org/anthology/2020.lrec-1.223",
    pages = "1803--1813",
    language = "English",
    ISBN = "979-10-95546-34-4",
}

Note that each LinCE dataset has its own citation too. Please see here for the correct citation on each dataset.

Contributions

Thanks to @lhoestq, @thomwolf, @gaguilar for adding this dataset.