Datasets:

austin2029
/

open-myanmar-data

language:
  - my
license: cc-by-4.0
tags:
  - myanmar
  - burmese
  - history
  - nlp
  - text-classification
  - rag
pretty_name: Open Myanmar Data - Myanmar Kings & Historical Narrative Dataset
dataset_info:
  features:
    - name: row_id
      dtype: string
    - name: source_topic
      dtype: string
    - name: sentence
      dtype: string
    - name: data_type
      dtype: string
  splits:
    - name: train
      num_bytes: 587678
      num_examples: 1242
      download_size: 174897
      dataset_size: 587678
  configs:
    - config_name: default
      data_files:
        - split: train
          path: data/train-*

🇲🇲 Open Myanmar Data: Myanmar Kings & Historical Narrative Dataset

📌 Dataset Summary

This repository contains a clean, high-quality, sentence-segmented historical narrative dataset regarding Myanmar history, royal dynasties (Pagan, Toungoo, Konbaung), and notable historical figures extracted directly from Myanmar Wikipedia.

Curated and cleaned by Aung Ko Ko Oo (Founder, Burmese NLP Research Group).

📊 Dataset Overview

Total Sentences: 1,242
Duplication Rate: 0.00% (100% unique clean sentences)
Data Type: Historical Narrative
Source: Myanmar Wikipedia (Attribution under CC BY-SA 4.0)

📂 Dataset Structure

Column	Data Type	Description
`row_id`	String	Unique identifier (e.g., `MY_HIST_00001`)
`source_topic`	String	Wikipedia topic/entity title
`sentence`	String	Clean Myanmar sentence ending with standard boundary (`။`)
`data_type`	String	Text category (`Historical Narrative`)

💡 Intended Use Cases

Retrieval-Augmented Generation (RAG): Knowledge indexing for Burmese historical QA chatbots.
Sentence Embedding & Vector Search: Benchmarking Burmese semantic similarity and clustering.
Information Extraction (IE): Named Entity Recognition (NER) and Relation Extraction for Burmese history.

💻 How to Use

1. Load via Hugging Face `datasets` Library

from datasets import load_dataset

# Load from Hugging Face Hub
dataset = load_dataset("austin2029/open-myanmar-data")
print(dataset['train'][0])

import pandas as pd

df = pd.read_csv("myanmar_kings/myanmar_kings_dataset.csv")
print(df.head())

@misc{open_myanmar_data_2026,
  author = {Aung Ko Ko Oo},
  title = {Open Myanmar Data: Myanmar Kings & Historical Narrative Dataset},
  year = {2026},
  publisher = {GitHub / Hugging Face},
  howpublished = {\url{[https://github.com/Aung-Ko-Ko-Oo/open-myanmar-data](https://github.com/Aung-Ko-Ko-Oo/open-myanmar-data)}}
}