No Detail Found!

Comparative analysis of classification methods of the dactyl alphabet of the Kazakh Language

Author(s)
NURZADA AMANGELDY Faculty of Information Technologies, L.N. Gumilyov Eurasian National University, Nur-Sultan 010008, Kazakhstan;
SAULE KUDUBAYEVA Faculty of Information Technologies, L.N. Gumilyov Eurasian National University, Nur-Sultan 010008, Kazakhstan;
BIBIGUL RAZAKHOVA Faculty of Information Technologies, L.N. Gumilyov Eurasian National University, Nur-Sultan 010008, Kazakhstan;
MUKANOVA ASSEL Higher School of Information Technology and Engineering, Astana International University, Nur-Sultan 010000,Kazakhstan
TURSYNOVA NAZIRA Faculty of Information Technologies, L.N. Gumilyov Eurasian National University, Nur-Sultan 010008, Kazakhstan;
Abstract
Kazakh Sign Language (KSL) is relatively new, and there is still no full-fledged system that can automatically recognize the Kazakh language's dactylic alphabet since Russian letters in the Kazakh alphabet have been identified until today. In order to prove that the Kazakh sign language can exist as a separate sign language, a comparison of the form of the display was carried out in terms of configuration (arm /forearm), place of execution (localization), the direction of movement, nature of movement and component that cannot be performed manually (facial expression and articulation) of Kazakh, Russian, English, Turkish sign languages. As a result of the study, it was proved that the Kazakh sign language can exist as a separate sign language by observing the forms of demonstration of 1050 words in 4 languages. Further, a comparative analysis of classification methods such as support vector machines, ensemble classifiers and nearest neighbor classifiers of dactyl alphabet letters was carried out to identify the optimal method.
Publication Details
Page(s) 5506-5513
DOI DOI not available
Published Journal: Journal of Theoretical and Applied Information Technology, Volume: 100, Issue: 19, Year: 2022

Request the full text from PASTIC

Keywords
support vector machines, recognition, Nearest Neighbor Classifiers Pattern, Palm Definition Model, Ensemble Classifiers, Sign Language, Mediapipe Hands, Multiple Classification, Hand Shape
0

Citations

0

Downloads

38

Views