Skip to content
Conference Open access

A Terminological Glossary and Standardization Platform for Uzbek Natural Language Processing

Sep 2026 · 2026 11th International Conference on Computer Science and Engineering (UBMK) · pp. 1427-1431 · 0 citations · 14 references

Abstract

The rapid growth of artificial intelligence and natural language processing has introduced many technical terms into Uzbek scientific, educational, and software-localization practice. In the absence of controlled terminology, the same concept may receive competing translations, unstable abbreviations, or inconsistent definitions. This paper presents a web-based terminological glossary and standardization platform for Uzbek natural language processing. The resource contains 927 unique terms in 22 thematic categories and represents each concept in Uzbek, English, Turkish, and Russian. Its principal contribution is not a new retrieval or embedding algorithm; it is an Uzbek-specific governance and dissemination layer that combines concept-oriented term records, evidence and source notes, expert voting, variant management, revision history, and machine-readable access. Standard web technologies and LaBSE-assisted similarity are used as implementation components. A 20-query, author-constructed pilot is reported only as a functional check: 15 exact queries returned the intended entry, three misspellings triggered the intended suggestion, and two out-of-vocabulary queries produced no entry. Because the sample is small and non-independent, no claim is made about general search effectiveness, duplicate-detection accuracy, user satisfaction, or expert agreement. The paper therefore specifies a systematic next-stage evaluation protocol based on held-out user queries, independent judgments, labeled duplicate pairs, agreement statistics, and API performance measurements.

Read PDF

We use cookies to run the site and, with your consent, for analytics and to show ads. See our Cookie Policy.