A Computational Decision-Support Framework for Analyzing Historical Sound Correspondences and Functional Lexical Evolution in Turkic Languages
Computer and Decision Making: An International Journal · 03.11.2026 · 3 (6S) · 1372-1384
DOI: 10.1922/comdem.v3i6s.18917 ↗Kurz gefasst
Fachartikel in der Zeitschrift Computer and Decision Making: An International Journal, online erschienen am 03.11.2026. Verfasst von Khulkharai Abdurakhmanova und 4 weiteren Autor:innen. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0).
Abstract (Originalsprache)
The historical sound correspondences and functional lexical variation are important for analyzing language families among the Turkic languages. An interpretable computational framework was put in place to explore historical and contemporary Turkic phonological features and similarities and cognate relationships. The analysis was based on the Savelyev–Robbeets Turkic lexical database which contains 8360 lexical items, 254 semantic concepts, 32 language varieties and 905 sets of cognates. The phonological feature extraction focused on the distribution of rhotic and lateral consonants, especially between vowels, and the functional lexical analysis on the study of comitative expressions. The normalized Levenshtein distance and sound-class based comparison were used to assess cross-language nearest-neighbor cognate retrieval. A total of 2123 rhotic and 1533 lateral consonant occurrences were identified, giving intervocalic frequencies of 26.24% and 32.81%, respectively. When examining 46 comitative lexical records, it is noticed that these words vary in structure – birle, birla, bilen, bulan, menen etc. Normalized Levenshtein distance obtained a correct cognate retrieval rate of 93.58% which is similar to that of sound-class comparison (93.35%). The results show how useful are the methods based on similarity and interpretable for systematic lexical comparison and for phonological analysis. Positional consonant frequencies, and lexical resemblance, themselves do not determine sound change direction. The proposed framework facilitates the reproducible comparative linguistic analysis and serves as a basis for further studies based on larger historical data, contextual phonological modelling and on independent validation procedures.
Zitieren
APA
Abdurakhmanova, K., Atakulova, M., Sagyndykova, R., Karataeva, G., & Tulanbaeva, Z. (2026). A Computational Decision-Support Framework for Analyzing Historical Sound Correspondences and Functional Lexical Evolution in Turkic Languages. Computer and Decision Making: An International Journal, 3(6S), 1372-1384. https://doi.org/10.1922/comdem.v3i6s.18917
Vancouver
Abdurakhmanova K, Atakulova M, Sagyndykova R, Karataeva G, Tulanbaeva Z. A Computational Decision-Support Framework for Analyzing Historical Sound Correspondences and Functional Lexical Evolution in Turkic Languages. Computer and Decision Making: An International Journal. 2026;3(6S):1372-1384. doi:10.1922/comdem.v3i6s.18917
BibTeX
@article{Abdurakhmanova2026,
author = {Abdurakhmanova, Khulkharai and Atakulova, Merim and Sagyndykova, Rakhat and Karataeva, Gulistan and Tulanbaeva, Zulfinaz},
title = {A Computational Decision-Support Framework for Analyzing Historical Sound Correspondences and Functional Lexical Evolution in Turkic Languages},
journal = {Computer and Decision Making: An International Journal},
year = {2026},
volume = {3},
number = {6S},
pages = {1372--1384},
doi = {10.1922/comdem.v3i6s.18917}
}
EUVIMED