DatarrX/tipitaka-dataset

Myanmar Tipitaka Dataset (Pali Corpus)

Dataset Summary

The Myanmar Tipitaka Dataset is a high-quality, structured collection of the Buddhist Pali Canon, transcribed in the Myanmar (Burmese) script. This dataset contains 462,504 paragraphs, covering the entire "Triple Basket" (Tipitaka) of Theravada Buddhism, including the original Mula (Canonical texts), Atthakatha (Commentaries), and Tika (Sub-commentaries).

This project was initiated by DatarrX to provide a clean, machine-readable corpus for Natural Language Processing (NLP) tasks, specifically focusing on the Pali language as preserved in Myanmar.

Project Credits & Governance

• Organization: DatarrX
• Lead Developer: Khant Sint Heinn (Kalix Louis)
• Source Data: pndaza/tipitaka-pali
• License: MIT License

---

What is the Tipitaka? (For Non-Religion Researchers)

The Tipitaka (Pali for "Triple Basket") is the standard collection of scriptures in the Theravada Buddhist tradition. It is one of the oldest and most extensive philosophical and linguistic records in human history.

1. Vinaya Pitaka: Rules and discipline for the monastic community.
2. Sutta Pitaka: Discourses and sermons of the Buddha (contains ethics, philosophy, and psychology).
3. Abhidhamma Pitaka: A detailed analytical and philosophical systematization of the Buddha's teachings.

For NLP researchers, this dataset represents a low-resource language corpus (Pali in Myanmar script) that is rich in complex grammar, ancient vocabulary, and logical structures.

---

Dataset Structure

The data is provided in JSONL (JSON Lines) format. Each line represents a paragraph or a verse with full hierarchical metadata.

Data Fields

Field	Description
id	Unique identifier (derived from source DB page ID and sequence).
category	An object containing the hierarchy: nikaya, basket, book, and chapter.
paragraph_no	The traditional paragraph numbering used in Buddhist scriptures.
content	Clean text content (HTML tags stripped).
metadata	Includes style (e.g., bodytext, gatha/verse) and db_page_id.

JSON Example

{
  "id": "db_00001_005",
  "category": {
    "nikaya": "ဝိနယပိဋက(ခုဒ္ဒကနိကာယ)",
    "basket": "mula",
    "book": "ပါရာဇိကပါဠိ",
    "chapter": "ဝေရဉ္ဇကဏ္ဍံ"
  },
  "paragraph_no": "၁",
  "content": "၁။ တေန သမယေန ဗုဒ္ဓေါ ဘဂဝါ ဝေရဉ္ဇာယံ ဝိဟရတိ...",
  "metadata": {
    "style": "bodytext",
    "db_page_id": 1
  }
}

---

Technical Specifications

• Total Records: 462,504
• Source Format: SQLite (converted to JSONL)
• Script: Myanmar Script
• Language: Pali
• Cleanliness: 100% success rate in extraction with zero skipped errors.

---

Potential Use Cases

This dataset is designed for a wide range of AI and Linguistic applications:

1. Large Language Models (LLM): Pre-training or fine-tuning models to understand ancient Pali logic or Myanmar script nuances.
2. Machine Translation: Building translation systems between Pali and Modern Burmese/English.
3. Dictionary/Search Engines: Creating semantic search tools for religious scholars.
4. Text Classification: Identifying styles (Poetry/Gatha vs. Prose/Bodytext) using the style metadata.
5. Digital Humanities: Analyzing the evolution of the Pali Canon across different Nikayas and Baskets.

---

Citation & Acknowledgments

If you use this dataset in your research, please cite the original data source and the DatarrX project:

Source Data: Pndaza (2024). Tipitaka Pali SQLite Database. Available at: https://github.com/pndaza/tipitaka-pali Dataset Curation: Khant Sint Heinn (Kalix Louis), DatarrX Organization (2026). Myanmar Tipitaka Dataset.

---

Contact

For inquiries or collaborations regarding Myanmar NLP projects, please reach out to DatarrX or visit our Hugging Face Profile.

---

About the Author

Khant Sint Heinn, working under the name Kalix Louis, is a Machine Learning Engineer focused on Natural Language Processing (NLP), data foundations, and open-source AI development. His work is centered on improving support for the Burmese (Myanmar) language in modern AI systems by building high-quality datasets, practical tools, and scalable infrastructure for language technology.

He is currently the Lead Developer at DatarrX, where he develops data pipelines, manages large-scale data collection workflows, and helps create open-source resources for researchers, developers, and organizations. His experience includes data engineering, web scripting, dataset curation, and building systems that support real-world machine learning applications.

Khant Sint Heinn is especially interested in advancing low-resource languages and making AI more accessible to underrepresented communities. Through his open-source contributions, he works to strengthen the Burmese (Myanmar) tech ecosystem and provide reliable building blocks for future language models, search systems, and intelligent applications.

His goal is simple: to turn limited language resources into practical opportunities through clean data, useful tools, and community-driven innovation.

Connect with the Author:
GitHub | Hugging Face | Kaggle