Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › Research › peer-review
Rule-Based Syntactic Analysis for Uzbek Language: An Alternative Approach to Overcome Data Scarcity and Enhance Interpretability. / Mengliev, Davlatyor B.; Barakhnin, Vladimir B.; Ibragimov, Bahodir B.
24th IEEE International Conference of Young Professionals in Electron Devices and Materials, EDM 2023; Novosibirsk; Russian Federation; 29 June 2023 до 3 July 2023. Institute of Electrical and Electronics Engineers (IEEE), 2023. p. 1910-1915.Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › Research › peer-review
}
TY - GEN
T1 - Rule-Based Syntactic Analysis for Uzbek Language: An Alternative Approach to Overcome Data Scarcity and Enhance Interpretability
AU - Mengliev, Davlatyor B.
AU - Barakhnin, Vladimir B.
AU - Ibragimov, Bahodir B.
PY - 2023/7
Y1 - 2023/7
N2 - This research paper introduces an innovative rule-based syntactic analysis algorithm specifically tailored for the Uzbek language, designed to address and overcome the challenges associated with insufficient data availability commonly faced in machine learning approaches. By leveraging the unique characteristics of Uzbek grammar, the study establishes a comprehensive rule set for effectively parsing sentences in this low-resource language. It covers various aspects including tokenization, construction of dependency trees, followed by rigorous optimization and testing performed on diverse Uzbek texts. Despite the absence of machine learning, the study's relevance is elevated by providing a solution to data scarcity, offering a transparent, interpretable system that ensures faster development, reduced computational resource requirements, and enhanced resilience to noise and errors in data. The paper provides a thorough examination of Uzbek grammar, syntactic features, and a set of parsing rules. It also reviews related works, outlines the proposed algorithm's development, and presents the future potential for Natural Language Processing techniques in low-resource languages like Uzbek.
AB - This research paper introduces an innovative rule-based syntactic analysis algorithm specifically tailored for the Uzbek language, designed to address and overcome the challenges associated with insufficient data availability commonly faced in machine learning approaches. By leveraging the unique characteristics of Uzbek grammar, the study establishes a comprehensive rule set for effectively parsing sentences in this low-resource language. It covers various aspects including tokenization, construction of dependency trees, followed by rigorous optimization and testing performed on diverse Uzbek texts. Despite the absence of machine learning, the study's relevance is elevated by providing a solution to data scarcity, offering a transparent, interpretable system that ensures faster development, reduced computational resource requirements, and enhanced resilience to noise and errors in data. The paper provides a thorough examination of Uzbek grammar, syntactic features, and a set of parsing rules. It also reviews related works, outlines the proposed algorithm's development, and presents the future potential for Natural Language Processing techniques in low-resource languages like Uzbek.
KW - Uzbek language
KW - rule-based
KW - syntactic analysis
KW - grammar
KW - tokenization
KW - dependency tree
KW - algorithm
KW - data scarcity
KW - transparency
KW - interpretability
KW - Natural Language Processing (NLP)
UR - https://www.scopus.com/record/display.uri?eid=2-s2.0-85171977695&origin=inward&txGid=260279ddddc7ab9a7cdece31fcce43b3
UR - https://www.mendeley.com/catalogue/efa98888-9ef8-3d74-b04d-6ef1fae6279c/
U2 - 10.1109/edm58354.2023.10225235
DO - 10.1109/edm58354.2023.10225235
M3 - Conference contribution
SN - 9798350336870
SP - 1910
EP - 1915
BT - 24th IEEE International Conference of Young Professionals in Electron Devices and Materials, EDM 2023; Novosibirsk; Russian Federation; 29 June 2023 до 3 July 2023
PB - Institute of Electrical and Electronics Engineers (IEEE)
ER -
ID: 55580573