kalixlouiis/lokaniti-vocabulary-pairs

Lokaniti Pali-Burmese Vocabulary Pairs

📖 About the Dataset

This dataset provides a clean, deduplicated collection of Pali and Burmese word/phrase pairs extracted directly from the Nissaya breakdowns of the Lokaniti text. It is designed to support lexicography, alignment tasks, and machine translation experiments involving classical Pali and Burmese languages.

• Total Unique Pairs: 1,640
• Data Integrity: 0 null values.

👤 Who Created

This dataset and its processing pipeline were independently created, curated, and maintained by Khant Sint Heinn (Kalix Louis).

🗂️ Dataset Structure

Each entry in the .jsonl file contains a direct word-to-word or phrase-to-phrase mapping with all punctuation removed:

{
  "pali": "အဟံ",
  "burmese": "ငါသည်"
}

🛠️ How It Was Created

The vocabulary pairs were automatically extracted and refined from the primary text via a processing script:

1. Parsing the traditional Burmese Nissaya commentary lines.
2. Isolating corresponding Pali tokens and Burmese gloss meanings.
3. Stripping all punctuation marks (၊, ။, -, etc.) to create clean lexical pairs.
4. Filtering out duplicate entries to ensure high dataset uniqueness.

The source code and data pipeline scripts are publicly available here:

👉 Lokaniti Dataset Pipeline GitHub Repository

🔗 Related Datasets

• Main Verse Dataset: lokaniti-verses-nissaya — Full verses, chapters, and multi-language translations.

📝 License & Citation

This dataset is open-source and released under the Creative Commons Attribution 4.0 International (CC-BY-4.0) license.

If you use this dataset, please cite it as:

@dataset{lokaniti_vocab_2026,
  author = {{Khant Sint Heinn}},
  title = {Lokaniti Pali-Burmese Vocabulary Pairs},
  year = {2026},
  publisher = {Hugging Face},
  url = {https://huggingface.co/datasets/kalixlouiis/lokaniti-vocabulary-pairs}
}

---

👤 About the Author

Khant Sint Heinn (ခန့်ဆင်ဟိဏ်း) | A.k.a. Kalix Louis
Founder & Lead AI Scientist @ DatarrX Foundation

I am a Machine Learning Engineer based in Myanmar, focused on Natural Language Processing (NLP) and Data-Centric AI. My work centers on building robust data foundations for low-resource languages, specifically Burmese (Myanmar)—by creating high-quality datasets, custom tokenizers, and specialized linguistic tools to democratize AI innovation.

Through open-source initiatives at DatarrX, I develop scalable data pipelines and computational resources—spanning contemporary Burmese corpora to specialized linguistic datasets, to provide reliable building blocks for next-generation language models and computational linguistics.

• 🌐 Portfolio: LINK!
• 🏢 Hugging Face: @kalixlouiis
• 🐙 GitHub: @kalixlouiis
• 💼 LinkedIn: Khant Sint Heinn
• 📧 Contact: kalixlouiis@gmail.com

If you find this project useful or are interested in collaborating on Burmese NLP, LLM evaluation, or low-resource language research, feel free to reach out, open an issue, or contribute!