Research and development of methods and algorithms for intellectual processing of ethnographic texts in the Kazakh language

The IRN AP13268833 project is carried out within the framework of grant funding for young scientists under the Zhas Galym project funded by the Science Committee of the Ministry of Science and Higher Education of the Republic of Kazakhstan.

Project Manager (postdoctoral student): K.E. Iklasova, Candidate of Medical Sciences, Associate Professor of the Department of Information and Communication Technologies (e-mail kiklasova1205@gmail.com , in ORCID: https://orcid.org/0000-0002-8330-4282 m, h-index = 3 (Scopus), Scopus author ID: 57209837609).

Scientific consultant (consultant): In Kulikova.P., Ph.D., Professor of the Department of Information and Communication Technologies (e-mail v4lentina@mail.ru , in ORCID: https://orcid.org/0000-0001-8198-2672 m, h-index = 1 (Scopus), Scopus author ID: 57209829779).

Terms of execution: 01.05.2022 - 31.12.24

Duration: 32 months.

Total amount of financing: 18,974,780 tenge.

The purpose of the project: To study existing and develop new methods and algorithms for intelligent data processing, namely, a Latin text in the Kazakh language for effective management of the transition to the Latin alphabet.

Expected results: Methods will be developed to support the transition to the Latin alphabet, providing semantic transliteration. This result is necessary for the evaluation of mechanical and semantic transliteration, for further implementation in the development of digital resources. To investigate the peculiarities of the formation of the subcorpus of the Kazakh language on Latin graphics (using the example of a terminological dictionary). This result is necessary for the development of algorithms and methods of intellectual processing of natural language and the creation of the basis for the formation of the corpus of the Kazakh language in Latin. To investigate/develop algorithms and methods using machine learning for semantic analysis of short Kazakh texts in Latin. This result is necessary for the practical assessment of resources and efficiency in software tools when extracting information from a natural language text presented in Latin graphics. Develop recommendations/algorithms to improve the efficiency of managing the transition to the Latin alphabet. This result is necessary for the formation of a strategy/policy for managing the process of transition of the Kazakh language script to Latin graphics.

Project description: One of the important tasks of the state program for the implementation of language policy in the Republic of Kazakhstan for 2020 - 2025 are: modernization of the writing of the Kazakh language based on the Latin alphabet, its improvement, unification and codification; improvement of the use of the Kazakh language in the field of informatization and communication. The solution of these problems is directly related to the research, development and application of information and communication technologies for processing large-volume digital resources of mass consumption, in Latin, which has not been sufficiently investigated. The idea of the project is to identify contradictions, weaknesses, inconsistencies in the processes of transition to Latin graphics, as well as to develop appropriate algorithms and methods of intellectual processing of Latinographic text in the Kazakh language that contribute to an effective transition to the Latin alphabet.

List of articles:

1. Yerkebulan G., Kulikova V., Kulikov V., Kulsharipova Z. (2021). Development of an entropy-based approach to identifying patterns in multilingual texts. Eastern European Journal of Corporate Technologies, 2 (2 (110)), 16-22. doi: https://doi.org/10.15587/1729-4061.2021.228695 (Percentile - 51).
2. Kulikov, V., Kulikova, V., & Yerkebulan, G. (2021). Detection of Google/Yandex translation in the multilingual text template identification system. International Journal of Computer Engineering, 20(1), 72-77. https://doi.org/10.47839/ijc.20.1.2094