Publications
Dataset · Zenodo · April 8, 2026
Uzbek 65 million web corpus lemmatized FastText model
Surayyo Khajibaeva, Khabibulla A. Madatov, Jernej Vičič
Abstract
FastText lemmatized model developed from 65 million Uzbek web corpus (10.5281/zenodo.19462612). The model can be used for efficient text representation via Word embeddings (vector representations). A simple Python script for loading and using the model. FastText is a lightweight, open-source library developed by Facebook AI Research for efficient text representation and classification. The FastText implementation used to produce the model, which can later be used for the model, was Gensim. The original corpus was lemmatized using UzbekLemma lemmatizer: https://pypi.org/project/UzbekLemma/.
Topic ModelingNatural Language Processing TechniquesSentiment Analysis and Opinion MiningUzbek languageword embeddingsFastText
More by Jernej Vičič
- AI-generated text detection for Slavic languagesZenodo · 2026
- Dataset of Uzbek base words: extraction and data analysis based on the school corpusSocial Sciences & Humanities Open · 2026 · Citations: 2
- Dataset of Uzbek base wordsRepository of the University of Primorsk · 2026
- Uzbek 65 million web corpus lemmatized Word2Vec modelZenodo · 2026