Skip to content
Publications

Dataset · Zenodo · August 22, 2026

UzbekLexLacuna: A Sense-Level Uzbek–English Lexical Lacuna Dataset

Dauletov Adilbek Yusupbayevich, Khamroeva Shahlo Mirdjonovna, Noila Matyakubova

Abstract

UzbekLexLacuna is a sense-level Uzbek–English lexical lacuna dataset developed for lexical-gap detection, word-sense analysis, machine translation evaluation, cross-lingual lexical semantics, and culturally sensitive NLP. The public release contains 860 benchmark-ready records, including 760 TRUE_LACUNA senses and 100 NOT_LACUNA polysemy-control senses. The dataset includes annotations for gap extent, gap source, English equivalence, meaning loss, translation strategy, provenance, and benchmark splits. The release also provides annotation documentation, controlled vocabularies, dataset statistics, and validation results.

Natural Language Processing TechniquesLexicography and Language StudiesTranslation Studies and PracticesUzbek languageUzbek-English translationword senses

More by Noila Matyakubova