EXPLAINABILITY OF THE SVM CLASSIFICATION MODEL FOR SENTIMENT ANALYSIS TASK OF UZBEK LANGUAGE
Keywords:
SVM, sentiment analysis, explainability, Uzbek languageAbstract
This paper investigates the integration of local model-agnostic explanations with support vector machine models to enhance explainability in sentiment analysis for the Uzbek language. While SVM models are effective for classification tasks, they often function as black-box models with limited transparency. To address this, we used LIME, which perturbs input data and observes changes in the model's output, revealing the text features that most influence classification. This approach improves transparency and trust in AI systems. Our case study focuses on sentiment analysis in the low-resource Uzbek language, showing how LIME aids in understanding SVM model decisions.
References
Cristianini N, Shawe-Taylor J. An Introduction to Support Vector Machines and Other Kernel-based Learning Methods. Cambridge: Cambridge University Press; 2000.
Marco Tulio Ribeiro, Sameer Singh, and Carlos Guestrin. 2016. "Why Should I Trust You?": Explaining the Predictions of Any Classifier. In Proceedings of the 22nd ACM SIGKDD International Conference on Knowledge Discovery and Data Mining (KDD '16). Association for Computing Machinery, New York, NY, USA, 1135–1144.
Scott M. Lundberg and Su-In Lee. 2017. A unified approach to interpreting model predictions. In Proceedings of the 31st International Conference on Neural Information Processing Systems (NIPS'17). Curran Associates Inc., Red Hook, NY, USA, 4768–4777.
Zednik, C. Solving the Black Box Problem: A Normative Framework for Explainable Artificial Intelligence. Philos. Technol. 34, 265–288 (2021).
Matlatipov S, Rahimboeva H, Rajabov J, Kuriyozov E (2022) Uzbek sentiment analysis based on local restaurant reviews. The international conference on agglutinative language technologies as a challenge of natural language processing (ALTNLP), 1–11
Kuriyozov, E., Matlatipov, S., Alonso, M.A., Gómez-Rodríguez, C. (2022). Construction and Evaluation of Sentiment Datasets for Low-Resource Languages: The Case of Uzbek. In: Vetulani, Z., Paroubek, P., Kubis, M. (eds) Human Language Technology. Challenges for Computer Science and Linguistics. LTC 2019. Lecture Notes in Computer Science(), vol 13212. Springer, Cham.
Sanatbek Gayratovich Matlatipov, Jaloliddin Rajabov, Elmurod Kuriyozov, and Mersaid Aripov. 2024. UzABSA: Aspect-Based Sentiment Analysis for the Uzbek Language. In Proceedings of the 3rd Annual Meeting of the Special Interest Group on Under-resourced Languages @ LREC-COLING 2024, pages 394–403, Torino, Italia. ELRA and ICCL.
Ilyos Rabbimov, Iosif Mporas, Vasiliki Simaki, and Sami Kobilov. 2020. Investigating the effect of emoji in opinion classification of uzbek movie review comments. In Speech and Computer, pages 435–445, Cham. Springer International Publishing
Ulugbek Salaev, Elmurod Kuriyozov, and Carlos Gómez-Rodríguez. 2022a. A machine transliteration tool between uzbek alphabets. CEUR Workshop Proceedings, 3315:42 – 50.
Ulugbek Salaev, Elmurod Kuriyozov, and Carlos Gómez-Rodríguez. 2022b. Simreluz: Similarity and relatedness scores as a semantic evaluation dataset for uzbek language. 1st Annual Meeting of the ELRA/ISCA Special Interest Group on Under-Resourced Languages, SIGUL 2022 - held in conjunction with the International Conference on Language Resources and Evaluation, LREC 2022 - Proceedings, page 199 – 206.
Ulugbek I. Salaev, Elmurod R. Kuriyozov, and Gayrat R. Matlatipov. 2023. Design and implementation of a tool for extracting uzbek syllables. Proceedings of the 2023 IEEE 16th International Scientific and Technical Conference Actual Problems of Electronic Instrument Engineering, APEIE 2023, page 1750 – 1755.
Khabibulla Madatov, Shukurla Bekchanov, and Jernej Vičič. 2023. Automatic detection of stop words for texts in uzbek language. Informatica, 47(2).
I M Rabbimov and S S Kobilov. 2020. Multi-class text classification of uzbek news articles using machine learning. Journal of Physics: Conference Series, 1546(1):012097.
