Modelling Subword Embeddings of Low-Resource Languages for the Digitalisation of Industry 4.0 Industrial Systems

Abstract

The Industry 4.0 paradigm relies on intelligent methods for analysing textual data such as technical documentation, event logs, and decision support materials. Low-resource languages, however, remain underrepresented due to limited training data and the absence of high-quality language models, a challenge that is especially pronounced for morphologically rich agglutinative languages. This paper investigates semantic vector representation modelling for low-resource languages in an Industry 4.0 context, using the Tatar language as a case study. A comparative analysis of Word2Vec and FastText architectures is conducted with an emphasis on modelling morphological productivity and contextual relationships. The results demonstrate that subword-based approaches using character n-grams significantly improve robustness to rare and unseen word forms and enhance lexical coverage, which is crucial for industrial applications. A reproducible embedding construction methodology is proposed, including heterogeneous corpus compilation, hyperparameter optimisation, and multi-metric quality evaluation. Experimental results show that customised FastText models achieve statistically significant improvements over baseline approaches. The trained models and source code are publicly available on the Hugging Face platform, and an interactive web interface supports testing and visualisation. The proposed approach is applicable to downstream industrial NLP tasks and transferable to other low-resource agglutinative languages.

More by Mullosharaf Arabov

• Parameter-Efficient Adaptation of Large Language Models to a Low-Resource Agglutinative Language: A Corpus, Datasets, and a Reproducible Benchmark for the Bashkir LanguagearXiv · 2026
• A comprehensive study of Tatar text classification methods: from corpus creation and Tatar2Vec embeddings to benchmark establishmentModeling and Analysis of Information Systems · 2026 · Citations: 1
• A Comprehensive Analysis of Arabic Natural Language Processing Research: Trends, Topic Evolution, and Research Gaps -- A Bibliometric and Topic-Based StudyarXiv · 2026
• Large Language Models for Low-Resource Languages: A Conceptual Framework for an Electronic Explanatory Dictionary of the Tajik LanguagearXiv · 2026