Update README.md

README.md

@@ -1,51 +1,50 @@
 ---
 license: cc-by-4.0
+language:
+- mnw
+- en
+- my
+tags:
+- dictionary
+- lexicon
+- multilingual
+- pali
+- mon-language
+- NLP
+- ai-training
+pretty_name: Mon-Eng-Mya-Pali Lexicon Dataset
+size_categories:
+- 10K<n<100K
+task_categories:
+- translation
 ---
-# 🇲🇲 🇬🇧 Mon-English-Burmese-Pali Dictionary Data for AI Training & Retrieval
 
-This repository contains a comprehensive dataset of the Mon language, structured specifically for Artificial Intelligence (AI) applications, Large Language Models (LLMs), Machine Translation, and Retrieval-Augmented Generation (RAG) systems.
+# 🇲🇲 🇬🇧 Mon-Eng-Mya-Pali Lexicon Dataset for AI
 
-ဖိုင်ဒေတာဝေါဟာရ ကေုာံ အဘိဓာန်ဘာသာမန် သွက်ဂွံဗ္တောန် AI (Training) ကေုာံ စကာပ္ဍဲ AI Applications နကဵု Format ဗီုပြင် JSON/JSONL။
+This dataset is a comprehensive, multilingual lexicon of the Mon language (ဘာသာမန်), paired with English, Burmese (Myanmar), and Pali equivalents. It is structured specifically for Natural Language Processing (NLP), Large Language Model (LLM) fine-tuning, Machine Translation, and Retrieval-Augmented Generation (RAG) applications.
 
----
-
-## 📂 Dataset Structure (ဗီုပြင်ဒေတာ)
-
-The dataset has been processed from multiple alphabetized CSV files into clean, AI-ready formats:
-
-* **`Combined_Mon_Dict_AI_Format.json`**: Suitable for Vector Databases (Pinecone, ChromaDB), Custom GPTs Knowledge Bases, and Semantic Search.
-* **`Mon_AI_Training_Data.jsonl`**: Formatted in Chat/Instruction format (`messages`: `system`, `user`, `assistant`) for LLM Fine-Tuning (OpenAI, LLaMA, Claude).
-* **`X_list_Eng_Mon.csv`**: Bidirectional English-to-Mon translation reference dataset.
+ဖိုင်ဒေတာဝေါဟာရ ကေုာံ အဘိဓာန်ဘာသာမန် (၄ ဘာသာ) သွက်ဂွံဗ္တောန် AI (Training) ကေုာံ စကာပ္ဍဲ AI Applications နကဵု Format ဗီုပြင် JSON/JSONL။
 
 ---
 
-## 📊 Data Fields (ကဏ္ဍဒေတာမပါလုပ်)
+## 📂 Dataset Description
 
-Each dictionary entry contains the following structured attributes:
+The dataset consists of standardized dictionary entries compiled from various alphabetized source files. Each entry maps a Mon word to its corresponding meanings across multiple languages, including linguistic metadata where available.
 
-| Field Name | Type | Description | Example |
-| :--- | :--- | :--- | :--- |
-| `Mon` | String | The core Mon word/phrase | က |
-| `Eng` | String | English translation / definition | fish |
-| `Mya` | String | Burmese (Myanmar) translation | ငါး |
-| `Pali` | String | Pali root/derivative word (if applicable) | မစ္ဆ |
-| `source_file`| String | Metadata showing the origin file | `Ka_list_Mon_Eng_Mya_Pali.csv` |
+### Data Formats Available:
+1. **`Combined_Mon_Dict_AI_Format.json`**: Standard JSON array containing structured objects, perfect for Vector Databases and semantic search.
+2. **`Mon_AI_Training_Data.jsonl`**: Standard conversational JSON Lines format (`messages` array with `system`, `user`, and `assistant` roles) optimized for direct LLM fine-tuning (e.g., OpenAI, LLaMA).
 
 ---
 
-## 🤖 AI Integration & Usage (ဗီုလဵုဂွံစကာ ပ္ဍဲ AI)
-
-### 1. For Custom GPTs (Knowledge Base)
-Upload `Combined_Mon_Dict_AI_Format.json` to your Custom GPT's **Knowledge** section. Use the following System Prompt:
-> "You are an expert Mon Language Translation AI. You have access to a complete Mon-English-Burmese-Pali dictionary file in JSON format. When a user queries a word in Mon, English, or Burmese, look up the exact or closest match from the data, and provide the comprehensive translation, including its Pali root if available."
+## 📊 Dataset Structure & Fields
 
-### 2. For LLM Fine-Tuning (JSONL Format)
-The `Mon_AI_Training_Data.jsonl` file follows standard conversational training structures:
+### Data Points Example (JSON)
 ```json
 {
-  "messages": [
-    {"role": "system", "content": "You are an expert Mon translator and lexicographer."},
-    {"role": "user", "content": "What is the translation and definition for the Mon word 'က'?"},
-    {"role": "assistant", "content": "The Mon word 'က' means 'fish' in English, 'ငါး' in Burmese, and 'မစ္ဆ' in Pali."}
-  ]
+  "Mon": "က",
+  "Eng": "fish",
+  "Mya": "ငါး",
+  "Pali": "မစ္ဆ",
+  "source_file": "Ka_list_Mon_Eng_Mya_Pali.csv"
 }