vmujadia's picture
Update README.md
d30d264 verified
metadata
language:
  - en
  - hi
  - bn
  - gu
  - kn
  - ml
  - mr
  - ur
  - te
  - ta
license: cc-by-nc-4.0
size_categories:
  - 1M<n<10M
task_categories:
  - translation
tags:
  - parallel corpora
  - domains
  - education
  - Communication skill
  - Computer science
  - Economics
  - Health
  - History
  - Law
  - Management
  - Marketing
  - Mathematics
  - Natural science
  - Political science
  - Psychology
  - Teaching methods
  - Textile
  - Tourism
  - health
dataset_info:
  features:
    - name: domain
      dtype: string
    - name: source_language
      dtype: string
    - name: target_language
      dtype: string
    - name: source_text
      dtype: string
    - name: target_text
      dtype: string
  splits:
    - name: train
      num_bytes: 1204929943
      num_examples: 1907018
    - name: test
      num_bytes: 17912807
      num_examples: 30958
  download_size: 529110440
  dataset_size: 1222842750
configs:
  - config_name: default
    data_files:
      - split: train
        path: data/train-*
      - split: test
        path: data/test-*

Dataset Card for Dataset Name

Domain Parallel Corpora for Indian Languages (bhashik-parallel-corpora-education) Education Domain

Dataset Description

  • Contains Human post edited parallel corpora for English and 10 Indian Langauges (Can be used as nway parallel)
  • Train (++) and Test (0.5K) separation for each language pair/domain

Dataset Size and Domains

Domain Language Pair Count (passages/sentences)
Natural science eng_Latn-hin_Deva 140999
Natural science eng_Latn-mar_Deva 191936
Natural science eng_Latn-kan_Knda 185491
Natural science eng_Latn-tel_Telu 223146
Natural science eng_Latn-mal_Mlym 84
Natural science eng_Latn-tam_Taml 38
Computer science eng_Latn-mar_Deva 80170
Computer science eng_Latn-hin_Deva 71507
Computer science eng_Latn-kan_Knda 75354
Computer science eng_Latn-tel_Telu 75388
Computer science eng_Latn-urd_Arab 332
Computer science eng_Latn-tam_Taml 11902
Law eng_Latn-tel_Telu 80165
Law eng_Latn-mar_Deva 81581
Law eng_Latn-tam_Taml 90
Law eng_Latn-kan_Knda 78245
Law eng_Latn-hin_Deva 735
Mathematics eng_Latn-mar_Deva 62881
Mathematics eng_Latn-tel_Telu 59373
Mathematics eng_Latn-kan_Knda 40625
Mathematics eng_Latn-hin_Deva 26259
Mathematics eng_Latn-tam_Taml 286
Mathematics eng_Latn-mal_Mlym 163
health eng_Latn-tam_Taml 12728
health eng_Latn-guj_Gujr 299
health eng_Latn-tel_Telu 34855
health eng_Latn-hin_Deva 11802
health eng_Latn-mar_Deva 1904
health eng_Latn-kan_Knda 135
health eng_Latn-mal_Mlym 202
health eng_Latn-ben_Beng 370
Teaching methods eng_Latn-kan_Knda 32670
Teaching methods eng_Latn-tel_Telu 31828
Teaching methods eng_Latn-hin_Deva 19065
Teaching methods eng_Latn-mar_Deva 32327
Teaching methods eng_Latn-tam_Taml 17
Teaching methods eng_Latn-mal_Mlym 527
Teaching methods eng_Latn-urd_Arab 3909
Management eng_Latn-mar_Deva 14521
Management eng_Latn-kan_Knda 11001
Management eng_Latn-tel_Telu 11041
Health eng_Latn-hin_Deva 715
Psychology eng_Latn-hin_Deva 12651
Psychology eng_Latn-tel_Telu 12771
Psychology eng_Latn-mar_Deva 12822
Psychology eng_Latn-kan_Knda 12570
Marketing eng_Latn-hin_Deva 4903
Marketing eng_Latn-tel_Telu 4971
Marketing eng_Latn-mar_Deva 4858
Marketing eng_Latn-kan_Knda 5099
Political science eng_Latn-tel_Telu 11978
Political science eng_Latn-mar_Deva 11935
Political science eng_Latn-kan_Knda 12001
Political science eng_Latn-hin_Deva 7876
Textile eng_Latn-tel_Telu 7609
Textile eng_Latn-mar_Deva 7637
Textile eng_Latn-kan_Knda 7630
History eng_Latn-tel_Telu 4902
History eng_Latn-mar_Deva 4919
History eng_Latn-kan_Knda 4941
Economics eng_Latn-tel_Telu 5087
Economics eng_Latn-mar_Deva 5093
Economics eng_Latn-kan_Knda 5061
Tourism eng_Latn-tel_Telu 12798
Tourism eng_Latn-mar_Deva 12795
Tourism eng_Latn-kan_Knda 12813
Communication skill eng_Latn-tel_Telu 8426
Communication skill eng_Latn-mar_Deva 8537
Communication skill eng_Latn-kan_Knda 8627
Total 1937976

Dataset Structure

  • File name contains source language and target language based on train and test splits
  • {'domain', 'source_language', 'target_language', 'source_text', 'target_text'}

Source Data

Educational Lectures

Details

  • Curated by: LTRC, IIIT Hyderabad, India
  • Funded by: MOE, GOI, India
  • Shared by: MT-NLP, LTRC, IIIT Hyderabad, India
  • Language(s) (NLP): eng_Latn, kan_Knda, mar_Deva, tel_Telu, hin_Deva, urd_Arab, mal_Mlym, guj_Gujr, tam_Taml, ben_Beng
  • License: [More Information Needed]
  • Paper: BhashaVerse : Translation Ecosystem for Indian Subcontinent Languages []
  • Demo: [More Information Needed]

Project Investigator

  • Prof. Dipti Misra Sharma, LTRC, IIIT Hyderabad

Data Curators

  • LTRC Language Experts

BibTeX:

@misc{mujadia2024bhashaversetranslationecosystem,
      title={BhashaVerse : Translation Ecosystem for Indian Subcontinent Languages}, 
      author={Vandan Mujadia and Dipti Misra Sharma},
      year={2024},
      eprint={2412.04351},
      archivePrefix={arXiv},
      primaryClass={cs.CL},
      url={https://arxiv.org/abs/2412.04351}, 
}

Dataset Card Authors [optional]

[More Information Needed]

Dataset Card Contact

[More Information Needed]