Unveiling the variance of Uzbek language: A rule-based algorithm for dialect recognition

Abstract

The study presents an algorithm for recognizing dialects of Uzbek language, which is one of the low-resource languages. The authors did research about related works in this field, and included results of brief comparative analysis in second sub-section of the manuscript. Moreover, during the study a rule-based algorithm for dialect detection was developed, which uses 100 000 words dictionaries for each dialect in the analysis process. This approach differs from existing solutions by its transparency in the analysis process of the words, which allows tracking each step of the algorithm. For more optimized performance of the algorithm, the authors divided the corpus of each dialect into dictionaries, each dialect has 28 dictionaries and thus 140 dictionaries were generated. Testing of the algorithm showed its high accuracy in identifying dialect words and processing mixed dialect entries.

More by Mukhriddin Eshkulov

• DC Pump Aggregates Database for Autonomous Photovoltaic Water-Supply SystemsZenodo · 2026
• Spectral and Optical Performance Optimization of Monocrystalline Semi-Transparent Photovoltaic Panels Integrated into GreenhousesInternational Siberian workshops and tutorials on electron devices and materials · 2026
• AI-Based Multi-Objective Optimization of Semi- Transparent Photovoltaic Greenhouses Considering Crop Light ConstraintsInternational Siberian workshops and tutorials on electron devices and materials · 2026
• Development of hybrid approach for named entity recognition in Uzbek language textPeerJ Computer Science · 2026 · Citations: 1