SEMANTIC ROLE LABELING (SRL) FOR THE DISAMBIGUATION OF NATURAL LANGUAGE PROCESSING (NLP) SYSTEMS IN LOW-RESOURCED LANGUAGES AS KYRGYZ LANGUAGE
Cognitive analyses of the language by humans cover all components of the language, currently this requires from the machine as well. And widely used languages such as English NLP machines almost reach the level of performing this analysis coherently as humans do. Which started in accordance with Cho...
Saved in:
| Main Authors: | , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Issyk-Kul State University
2024
|
| Subjects: | |
| Online Access: | http://ir.unimas.my/id/eprint/47273/1/SEMANTIC%20ROLE%20LABELING%20%28SRL%29%20FOR%20THE.pdf http://ir.unimas.my/id/eprint/47273/ https://www.elibrary.ru/item.asp?id=75095075 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| Summary: | Cognitive analyses of the language by humans cover all components of the language, currently this requires from the machine as well. And widely used languages such as English NLP machines almost reach the level of performing this analysis coherently as humans do. Which started in accordance with Chomsky's Universal grammar theory to analyze language via algorithms which were implemented according to the language syntactical structure. However, experience has shown that there is a linguistic features which algorithm based just on language structure can’t analyze accurately as humans does. One of the reasons for this is that every word can have various grammatical and semantic features according to its particular context. Specifically in the case of languages with flexible word order. So NLP machines in such morphologically rich agglutinative languages such as Kyrgyz, need to cover both nature (grammatical, semantical) of the word to enhance the accuracy level of the tool. Thus this work will analyze the word’s semantic meaning beside grammatical features such as word correlation (who did what to whom, when and where). Thus this paper will investigate the challenges of implementing Semantic Role Labeling (SRL) in the context of the Kyrgyz language, a low-resourced language, by examining its unique linguistic properties and the limitations of existing NLP tools. With identification and analysis of the specific challenges faced by the SRL model in handling ambiguous or complex sentences in the Kyrgyz language, providing insights into areas for future improvements. |
|---|