kalixlouiis/myanmar-linguistic-ambiguitie-001

🇲🇲 Myanmar Linguistic Ambiguities - Dataset 001

A comprehensive corpus designed for the disambiguation and grammatical error correction of homophones and confusing particles in the Burmese language.

📚 Overview: Myanmar Linguistic Ambiguities Series

The Myanmar Linguistic Ambiguities (MLA) project is an ongoing effort to build highly precise and contextually rich datasets targeting specific common grammatical and orthographic errors in Burmese (Myanmar Language) writing. These errors often involve words that sound identical (homophones) but carry distinct grammatical functions, leading to significant challenges for downstream NLP tasks and Large Language Models (LLMs).

MLA-Dataset-001 focuses on the primary distinction between the two most frequently confused particles: လဲ and လည်း.

🎯 Dataset 001: The လဲ vs. လည်း Distinction

Dataset 001 provides 1600 unique data points meticulously labeled for contextual correctness, helping models discern the subtle grammatical roles of these two characters.

Key Distinction:

• လဲ: Primarily used as a Question Particle at the end of a sentence (e.g., "Where are you going?"). It is also used in words meaning Verb (to fall, to exchange, to change) or as part of Fixed Noun/Adverbial phrases (e.g., Pearl, Repeatedly).
• လည်း: Primarily used as an Inclusion Particle (Conjunction) meaning "also," "too," or "as well" (e.g., "I am going also."). It is also essential in fixed formal conjunctions (e.g., သော်လည်းကောင်း - "whether X or Y").

📊 Data Statistics

Category	Description	Data Points	Correctness Ratio (0:1)
Correct Use (1)	Contextually appropriate usage	1000	N/A
Incorrect Use (0)	Grammatically incorrect usage	600	N/A
Total Corpus Size		1600	5:3

🛠️ Data Structure and Labeling

The dataset is provided in a straightforward CSV format, making it instantly usable for classification, sequence labeling, or fine-tuning Burmese LLMs.

Column Name	Data Type	Description
sentence_id	Integer	Unique identifier for the data point (1 to 1600).
text	String	The complete Burmese sentence containing the target_word. (This often includes grammatical errors for incorrect instances).
target_word	String	The specific instance of 'လဲ' or 'လည်း' being evaluated in the sentence.
correct_label	String	The grammatically correct form for the context (i.e., 'လဲ' or 'လည်း').
is_correct_in_context	Boolean (0 or 1)	1 if target_word is correctly used; 0 if the usage is incorrect and should be corrected to correct_label.
rule_type	String	Categorization based on the grammatical function being tested (e.g., Question_Particle, Inclusion_Particle, Verb_Exchange, Conj(Incorrect).

Example Records (Conceptual)

sentence_id	text	target_word	correct_label	is_correct_in_context	rule_type
150	မင်း ဘယ်သွားမလည်း။	လည်း	လဲ	0	Question_Particle (Incorrect)
528	မင်းမှာ ပစ္စည်းရှိသလို ငါ့မှာလည်း ပစ္စည်းရှိသည်။	လည်း	လည်း	1	Inclusion_Particle
826	ဒီကားရဲ့ အင်ဂျင်ကို အသစ်နဲ့ လဲ ရမယ်။	လဲ	လဲ	1	Verb_Exchange
1401	ပညာရေးတွင်လဲကောင်း ကျန်းမာရေးတွင်လဲကောင်း အသုံးစရိတ်တိုးမြှင့်ရမည်။	လဲ	လည်း	0	Conj(Incorrect)

✍️ Data Creation Methodology

The MLA-Dataset-001 was manually constructed and labeled based on the official definitions provided in the Myanmar Language Commission's dictionary and standardized grammar rules.

1. Rule Identification: All primary and secondary grammatical functions of လဲ and လည်း were identified and categorized into distinct rule types.
2. Synthetic Sentence Generation: Contextually rich Burmese sentences were synthetically generated by the project maintainers, ensuring wide linguistic coverage and variety in subjects, tenses, and sentence structures.
3. Error Injection & Labeling: For the "Incorrect Use" categories, sentences were specifically created by injecting the incorrect homophone (လဲ instead of လည်း, or vice versa), and the true correct_label was applied.
4. Verification: All 1600 data points were double-checked to ensure consistency between the rule_type and the correct_label.

🚀 Future Datasets in the MLA Series

This dataset is the first installment in the series, targeting common Burmese orthographic challenges. Future releases will include:

• MLA-Dataset-002 (Coming Soon): Focusing on the distinction between the numeral particle တ and the numeral တစ်.
• MLA-Dataset-003 (Coming Soon): Focusing on the verb and particle distinction between ကျ and ကြ.

📜 License

This dataset is released under the Creative Commons Attribution-ShareAlike 4.0 International (CC BY-SA 4.0) license.

Usage Rights:

You are free to:

• Share — copy and redistribute the material in any medium or format.
• Adapt — remix, transform, and build upon the material for any purpose, even commercially.

Terms:

• Attribution — You must give appropriate credit, provide a link to the license, and indicate if changes were made. You may do so in any reasonable manner, but not in any way that suggests the licensor endorses you or your use.
• ShareAlike — If you remix, transform, or build upon the material, you must distribute your contributions under the same license as the original.

Please cite this repository (kalixlouiis/myanmar-linguistic-ambiguities-001) if you use this data in your research or applications.