Pali Text Dataset

About Dataset

The Pali Buddhist Verse Classification Dataset is a structured corpus developed for computational analysis and classification of Buddhist literature written in the Pali language. The dataset contains verses from the Theragatha and Therigatha collections of the Buddhist Canon (Tipitaka), traditionally attributed to male and female Buddhist saints respectively. The corpus includes both Devanagari and Roman script representations, enabling comparative studies of script-based language processing. The dataset contains 1,793 labeled verses, comprising 1,272 Theragatha verses and 521 Therigatha verses. The original XML files were converted to UTF-8 encoded CSV format. Pali verses in Devanagari script, Roman transliteration, and class labels were extracted, and numeric labels were mapped to descriptive class names (Theragatha and Therigatha). The citation of the dataset - Nikita Neveditsin, Ambuja Salgaonkar, Pawan Lingras, and Vijay Mago. (2024). “Classification of Buddhist Verses: The Efficacy and Limitations of Transformer-Based Models.” In M. Hämäläinen, E. Öhman, S. Miyagawa, K. Alnajjar, & Y. Bizzoni (Eds.), Proceedings of the 4th International Conference on Natural Language Processing for Digital Humanities (NLP4DH 2024), pp. 377–385. Association for Computational Linguistics. https://doi.org/10.18653/v1/2024.nlp4dh-1.37. This dataset was identified and facilitated for onboarding as part of the Dataset Onboarding Support Team (DOST) initiative led by CivicDataLab (CDL), partnering with the Gates Foundation in collaboration with BHASHINI. CivicDataLab facilitated the dataset discovery, validation, metadata preparation and onboarding process. All dataset ownership and intellectual property rights remain with the original author(s).