Aspect-Based Sentiment Analysis for Turkish Reviews Using Token and Sequential Classification Methods
DOI:
https://doi.org/10.24203/2drscx54Keywords:
Aspect Extraction, Deep Learning, Token ClassificationAbstract
Aspect-Based Sentiment Analysis (ABSA) aims to identify sentiments expressed toward specific aspects or attributes of entities in text. This study addresses the under-explored area of ABSA in the Turkish language by extracting aspect terms (targets) and their categories from customer reviews and determining the sentiment polarity for each aspect. Turkish, being a morphologically rich and structurally complex language, poses unique challenges that often hinder the direct application of methods developed for other languages. Hence, developing sentiment analysis approaches tailored to Turkish is of significant importance. We propose a two-stage pipeline: a token-level classification to recognize aspect terms and assign them to one of 12 predefined aspect categories, followed by a sequence-level (sentence-level) classification to predict sentiment (positive, negative, or neutral) for each identified aspect. We fine-tuned five transformer-based language models (BERT, ConvBERT, ELECTRA, DeBERTa, and DistilBERT) for aspect term and category extraction, and four models (BERT, ConvBERT, ELECTRA, DistilBERT) for sentiment classification. Experimental results on the SemEval-2016 Turkish ABSA Restaurant dataset show that the BERT model achieved the highest accuracy (92.20%) for aspect term and category identification, closely followed by ConvBERT (91.68%). For sentiment analysis, ConvBERT performed best with an accuracy of 86.91%, outperforming ELECTRA (85.34%), BERT (82.75%), and DistilBERT (77.48%). These findings demonstrate that pre-trained transformer models can effectively handle fine-grained sentiment analysis in Turkish, substantially improving on previous approaches. The proposed pipeline and comparative results provide a novel benchmark for Turkish ABSA, with potential applications in analyzing Turkish customer feedback to glean actionable insights.
References
[1] B. Kama, M. Ozturk, P. Karagoz, I. H. Toroslu, and O. Ozay, "A web search enhanced feature extraction method for aspect-based sentiment analysis for Turkish informal texts," in Big Data Analytics and Knowledge Discovery: 18th International Conference, DaWaK 2016, Porto, Portugal, vol. 18, Springer, 2016, pp. 225–238.
[2] B. Kama, M. Ozturk, P. Karagoz, I. H. Toroslu, and M. Kalender, "Analyzing implicit aspects and aspect dependent sentiment polarity for aspect-based sentiment analysis on informal Turkish texts," in Proceedings of the 9th International Conference on Management of Digital EcoSystems, 2017, pp. 134–141.
[3] J. Devlin, M.-W. Chang, K. Lee, and K. Toutanova, “BERT: Pre-training of Deep Bidirectional Transformers for Language Understanding,” arXiv.org, Oct. 11, 2018. https://arxiv.org/abs/1810.04805
[4] V. Sanh, L. Debut, J. Chaumond, and T. Wolf, "DistilBERT, a distilled version of BERT: smaller, faster, cheaper and lighter," arXiv preprint arXiv:1910.01108, 2019.
[5] Z. H. Jiang, W. Yu, D. Zhou, Y. Chen, J. Feng, and S. Yan, "ConvBERT: Improving BERT with span-based dynamic convolution," Advances in Neural Information Processing Systems, vol. 33, pp. 12837–12848, 2020.
[6] K. Clark, M. T. Luong, Q. V. Le, and C. D. Manning, "Electra: Pretraining text encoders as discriminators rather than generators," arXiv preprint arXiv:2003.10555, 2020.
[7] P. He, X. Liu, J. Gao, and W. Chen, "Deberta: Decoding-enhanced BERT with disentangled attention," arXiv preprint arXiv:2006.03654, 2020.
[8] T. Wilson, J. Wiebe, and P. Hoffmann, "Recognizing contextual polarity in phrase-level sentiment analysis," in Proceedings of Human Language Technology Conference and Conference on Empirical Methods in Natural Language Processing, 2005, pp. 347–354.
[9] K. Cho, B. Van Merriënboer, C. Gulcehre, D. Bahdanau, F. Bougares, H. Schwenk, and Y. Bengio, "Learning phrase representations using RNN encoder-decoder for statistical machine translation," arXiv preprint arXiv:1406.1078, 2014.
[10] Y. Wang, A. Sun, J. Han, Y. Liu, and X. Zhu, “Sentiment Analysis by Capsules,” Proceedings of the 2018 World Wide Web Conference on World Wide Web - WWW ’18, 2018, doi: https://doi.org/10.1145/3178876.3186015.
[11] X. Hou, J. Huang, G. Wang, X. He, and B. Zhou, "Selective attention based graph convolutional networks for aspect-level sentiment classification," arXiv preprint arXiv:1910.10857, 2019.
[12] M. Pontiki, D. Galanis, J. Pavlopoulos, H. Papageorgiou, I. Androutsopoulos, and S. Manandhar, "SemEval-2016 Task 5: Aspect-based sentiment analysis," in Proceedings of the 10th International Workshop on Semantic Evaluation (SemEval-2016), 2016, pp. 19–30.
[13] D. Zhang, Z. Zhu, Q. Lu, H. Pei, W. Wu, and Q. Guo, "Multiple interactive attention networks for aspect-based sentiment classification," Applied Sciences, vol. 10, p. 2052, 2020.
[14] E. H. d. Silva and R. M. Marcacini, "Aspect-based sentiment analysis using BERT with disentangled attention," in Proceedings, 2021.
[15] H. Yang and K. Li, "Improving implicit sentiment learning via local sentiment aggregation," arXiv preprint arXiv:2110.08604, 2021.
[16] K. Scaria, H. Gupta, S. A. Sawant, S. Mishra, and C. Baral, "InstructABSA: Instruction learning for aspect based sentiment analysis," arXiv preprint arXiv:2302.08624, 2023.
[17] H. Türkmen, S. İ. Omurca, and E. Ekinci, "An aspect based sentiment analysis on Turkish hotel reviews," Girne American University Journal of Social and Applied Sciences, vol. 6, pp. 12–15, 2016.
[18] E. Ekinci, H. Türkmen, and S. İ. Omurca, "Multi-word aspect term extraction using Turkish user reviews," International Journal of Computer Engineering and Information Technology, vol. 9, p. 15, 2017.
[19] F. S. Çetin and G. Eryiğit, "Türkçe hedef tabanlı duygu analizi için alt görevlerin incelenmesi – hedef terim, hedef kategori ve duygu sınıfı belirleme," Bilişim Teknolojileri Dergisi, vol. 11, pp. 43–56, 2018, doi:10.17671/gazibtd.325865.
[20] P. Karagoz, B. Kama, M. Ozturk, I. H. Toroslu, and D. Canturk, "A framework for aspect based sentiment analysis on Turkish informal texts," Journal of Intelligent Information Systems, vol. 53, pp. 431–451, 2019.
[21] D. Ozkan, Aspect-based sentiment analysis in Turkish, M.S. thesis, Atılım University, 2020.
[22] M. U. Salur, İ. Aydın, and M. Jamous, "An ensemble approach for aspect term extraction in Turkish texts," Pamukkale Üniversitesi Mühendislik Bilimleri Dergisi, vol. 28, pp. 769–776, 2021.
[23] A. B. Girgin, G. Gümüşçekiççi, and N. C. Birdemir, "Turkish sentiment analysis: A comprehensive review," Sigma Journal of Engineering and Natural Sciences, vol. 42, no. 4, pp. 1292–1314, 2023.
[24] P. Meesad, "Thai fake news detection based on information retrieval, natural language processing and machine learning," SN Computer Science, vol. 2, p. 425, 2021.
[25] J. C. Pereira-Kohatsu, L. Quijano-Sánchez, F. Liberatore, and M. Camacho-Collados, "Detecting and monitoring hate speech in Twitter," Sensors, vol. 19, p. 4654, 2019.
[26] G. Salton, R. J. Ross, and J. D. Kelleher, "Idiom token classification using sentential distributed semantics," unpublished.
[27] T. Wang, R. Sridhar, D. Yang, and X. Wang, "Identifying and mitigating spurious correlations for improving robustness in NLP models," arXiv preprint arXiv:2110.07736, 2021.
[28] C. Li, X. Peng, H. Peng, J. Li, and L. Wang, "TextGTL: Graph-based transductive learning for semi-supervised text classification via structure-sensitive interpolation," in Proceedings of the International Joint Conference on Artificial Intelligence (IJCAI), 2021, pp. 2680–2686.
[29] P. L. Prasanna and D. R. Rao, "Text classification using artificial neural networks," International Journal of Engineering & Technology, vol. 7, pp. 603–606, 2018.
[30] K. Zhao, L. Huang, R. Song, Q. Shen, and H. Xu, "A sequential graph neural network for short text classification," Algorithms, vol. 14, p. 352, 2021.
Downloads
Published
Issue
Section
License
Copyright (c) 2025 Metin Bilgin, Melek Turan
This work is licensed under a Creative Commons Attribution-NonCommercial 4.0 International License.
The articles published in International Journal of Computer and Information Technology (IJCIT) is licensed under a Creative Commons Attribution-NonCommercial 4.0 International License.
