COMPARATIVE ANALYSIS OF MACHINE LEARNING ALGORITHMS FOR AUTOMATIC TOPIC CLASSIFICATION OF KARAKALPAK NEWS TEXTS
Abstract
This paper investigates automatic topic classification of Karakalpak news texts. The experiment was conducted using the KKNEWS dataset, which initially contained 13,828 records. After analyzing the distribution of topic categories, seven categories containing at least 200 texts each were selected. Following script unification and duplicate removal, the final experimental corpus consisted of 13,442 Karakalpak news texts written in Latin script. TF-IDF was used for numerical text representation, while Logistic Regression, Linear Support Vector Machine, and Multinomial Naive Bayes were evaluated as classifiers. Linear SVM achieved the best performance with an accuracy of 0.8702 and a macro-F1 score of 0.7485. The results indicate that the combination of TF-IDF and Linear SVM can serve as an effective baseline approach for automatic topic classification of Karakalpak news texts.Keywords
Karakalpak language, natural language processing, text classification, machine learning, TF-IDF, Linear SVM, Logistic Regression, Naive BayesReferences
Madatov, K., Bekchanov, Sh., & Vičič, J. (2023). Dataset of Karakalpak language stop words. Data in Brief. 2023. Vol. 48. Article 109111. DOI: 10.1016/j.dib.2023.109111. https://doi.org/10.1016/j.dib.2023.109111
Abdurakhmonova, N. Z., Sayfullaeva, R. R., & Shirinova, R. K. (2025). Leveraging mBERT with Rule-Driven Pre-Processing for Domain-Specific Karakalpak Text Classification. International Conference on Computer Science and Engineering (UBMK), 2025. P. 311–314. DOI: 10.1109/UBMK67458.2025.11207028. 10.1109/UBMK67458.2025.11207028. https://doi.org/10.1109/UBMK67458.2025.11207028
Manning, C. D., Raghavan, P., & Schütze, H. (2008). Introduction to Information Retrieval. Cambridge: Cambridge University Press, 2008.
Joachims, T. (1998). Text Categorization with Support Vector Machines: Learning with Many Relevant Features. Proceedings of the European Conference on Machine Learning (ECML). Springer, 1998. P. 137–142. 137–142.
Devlin, J., & Chang, M. (2019). -W., Lee K., Toutanova K. BERT: Pre-training of Deep Bidirectional Transformers for Language Understanding. Proceedings of NAACL-HLT. 2019. P. 4171–4186. DOI: 10.18653/v1/N19-1423. 10.18653/v1/N19-1423. https://doi.org/10.18653/v1/N19-1423
Conneau, A., Khandelwal, K., Goyal, N., et al. (2020). Unsupervised Cross-lingual Representation Learning at Scale. Proceedings of the 58th Annual Meeting of the Association for Computational Linguistics. 2020. P. 8440–8451. 8440–8451.
Pedregosa, F., Varoquaux, G., Gramfort, A., et al. (2011). Scikit-learn: Machine Learning in Python. Journal of Machine Learning Research. 2011. Vol. 12. P. 2825–2830. Machine Learning in Python.
Joulin, A., Grave, E., Bojanowski, P., & Mikolov, T. (2017). Bag of Tricks for Efficient Text Classification. Proceedings of EACL. 2017. P. 427–431. 427–431.
Downloads
Published
How to Cite
Issue
Section
Pages
Citation
