An umbrella portal for all the glossary creation workflows and datasets
Indic-Glossaries is an open source portal that is an umbrella to expose all the workflows for the creation of Indic glossaries and also the created/curated glossary datasets.
The datasets are licensed under :
(The entire collected glossary datasets are submitted to ULCA platform. Please refer https://bhashini.gov.in/ulca for further details)
| Collection Source | Glossary Corpus Count | Download link |
|---|---|---|
| IndoWordNet | 1,724,816 | download |
| Bharatavani | 866,423 | download |
| CSTT | 319,420 | download |
| OSF | 212,951 | download |
| NLPC - Univ of Moratuwa | 19,273 | download |
| Anuvaad | 1,690 | download |
| NCF-NCERT | 272 | download |
| Language Pair | Glossary Corpus Count |
|---|---|
| English-Assamese | 75,256 |
| English-Bengali | 102,855 |
| English-Bodo | 158,457 |
| English-Dogri | 8,624 |
| English-Goan Konkani | 87,300 |
| English-Gujarati | 168,343 |
| English-Hindi | 852,324 |
| English-Kannada | 156,777 |
| English-Kashmiri | 69,673 |
| English-Maithili | 8,297 |
| English-Malayalam | 98,092 |
| English-Manipuri | 7,183 |
| English-Marathi | 106,787 |
| English-Nepali | 65,795 |
| English-Odia | 130,801 |
| English-Punjabi | 152,081 |
| English-Sanskrit | 130,042 |
| English-Sindhi | 4,797 |
| English-Tamil | 316,976 |
| English-Telugu | 92,839 |
| English-Urdu | 79,253 |
| Hindi-Tamil | 113 |
| Hindi-Telugu | 13,340 |
| Hindi-Urdu | 3,133 |
| Hindi-English | 130,422 |
| Malayalam-English | 9,269 |
| Sanskrit-Hindi | 111,155 |
| Tamil-Hindi | 4,868 |
| Domain | Glossary Corpus Count |
|---|---|
| general | 2,209,449 |
| economy | 21,762 |
| technology | 26,360 |
| education | 351,387 |
| geography | 92,169 |
| legal | 46,870 |
| financial | 17,163 |
| automobile | 6,303 |
| healthcare | 278,436 |
| national-security-and-defence | 30,570 |
| agriculture | 4,807 |
| parliamentary | 22,180 |
| history | 23,955 |
| news | 7,303 |
| lifestyle | 3,506 |
| entertainment | 1,143 |
| philosophy | 1,489 |
The goal is to build high quality glossary datasets for the Indian languages across various domains (General, Legal, Education, Healthcare, Automobile, News etc).
Read more about Glossary Explorer @ https://glossary.ai4bharat.org
Any information/help/discussion required, can be taken up using the following link : https://github.com/AI4Bharat/Indic-Glossaries/discussions
This project adheres to the Contributor Covenant code of conduct. By participating, you are expected to uphold this code. Please report unacceptable behavior to opensource@ai4bharat.org.