SAHRAN: Sentiment Analysis of Hotel Reviews with Attention-Based Recurrent Neural Network

Öz

Automatically analysing the sentiment of comments expressed by a user on a web page for any purpose is a rapidly expanding important research area. Text sentiment analysis, as it is known in the literature, is a technique that allows users to determine their emotional tendencies in comments defined for any purpose. Users comment on the content of web pages used by thousands of people such as vacation sites, shopping pages, social media, brand reviews, financial reviews, health sites, political pages. The comments made have the ability to directly affect a user who wants to benefit from these services in any way. For these reasons, it is important to examine people's emotions in their comments in automatic review of comments. Recurrent Neural Network (RNN) based architectures have achieved remarkable success in solving Natural Language Processing (NLP) problems. In this article, an RNN based deep learning model is proposed that works on a publicly available dataset obtained from the TripAdvisor web page and performs sentiment analysis. The proposed SAHRAN model uses an attention mechanism based on the dot product structure to capture emotional words in user comments. In the model, Bidirectional Gated Recurrent Unit (BiGRU) and Bidirectional Long Short Term Memory (BiLSTM) deep learning layers are integrated into the model to capture emotional features. As a result of the experimental studies, the proposed SAHRAN model achieved performance values of 0.9524, 0.9685, 0.9082 and 0.9338 in terms of precision, recall, F1 score and accuracy performance measures, respectively.

Anahtar Kelimeler

• RNN
• Deep learning
• BiGRU
• BiLSTM
• Natural language process

SAHRAN: Dikkat Tabanlı Yinelemeli Sinir Ağı ile Hotel Yorumlarının Duygu Analizi

Öz

Bir kullanıcının herhangi bir amaç içerisinde bulunan web sayfasında ifade edeceği yorumları otomatik olarak duygu yönünden analiz etmek hızla genişleyen önemli bir araştırma alanıdır. Literatürdeki adıyla metin duygu analizi, herhangi bir amaç ile tanımlanan yorumlardaki kullanıcıların duygusal eğilimlerini belirleyebilmeyi sağlayan bir tekniktir. Tatil siteleri, alışveriş sayfaları, sosyal medya, marka yorumları, finans yorumları, sağlık siteleri, siyaset sayfaları gibi binlerce insanın faydalandığı web sayfalarındaki içeriklerin kullanıcılar tarafından yorumlanması gerçekleştirilmektedir. Yapılan yorumlar, herhangi bir şekilde bu hizmetlerden faydalanmak isteyen bir kullanıcıyı doğrudan etkileme özelliğine sahiptir. Bu sebeplerden dolayı yorumların otomatik incelenmesinde insanların yorumlarındaki duygularını incelemek önem arz etmektedir. Yinelemeli Sinir Ağı (RNN) tabanlı mimariler Doğal Dil İşleme (NLP) problemlerinin çözümünde dikkat çekici başarılar sağlamıştır. Bu makale kapsamında tripadvisor web sayfasından elde edilen halka açık bir veriseti üzerinde çalışıp duygu analizi gerçekleştiren RNN tabanlı bir derin öğrenme modeli önerilmiştir. Önerilen SAHRAN modeli, kullanıcı yorumlarındaki duygusal sözcükleri yakalayabilmek için nokta çarpım yapısını temel alan bir dikkat mekanizması kullanılmıştır. Modelde, duygu özelliklerini yakalayabilmek için de Çift Yönlü Kapılı Yinelemeli Hücreler (BiGRU) ve Çift Yönlü Uzun Kısa Süreli Bellek (BiLSTM) derin öğrenme katmaları modele entegre edilmiştir. Yapılan deneysel çalışmalar neticesinde önerilen SAHRAN modeli hassasiyet, geri çağırma, F1 puanı ve doğruluk performans ölçütleri açısından sırasıyla 0.9524, 0.9685, 0.9082 ve 0.9338 performans değerlerini elde etmiştir.

Anahtar Kelimeler

• RNN
• Derin öğrenme
• BiGRU
• BiLSTM
• Doğal dil işleme

Kaynakça

1. Al Hamoud, A., Hoenig, A., & Roy, K. (2022). Sentence subjectivity analysis of a political and ideological debate dataset using LSTM and BiLSTM with attention and GRU models. Journal of King Saud University - Computer and Information Sciences, 34(10), 7974–7987.
2. Alam, M. H., Ryu, W.-J., & Lee, S. (2016). Joint multi-grain topic sentiment: modeling semantic aspects for online reviews. Information Sciences, 339, 206–223.
3. Aravinthan, A., & Eugene, C. (2024). Exploring Recent NLP Advances for Tamil: Word Vectors and Hybrid Deep Learning Architectures. The International Journal on Advances in ICT for Emerging Regions, 17(2).
4. Bali, A. P. S., Fernandes, M., Choubey, S., & Goel, M. (2019). Comparative Performance of Machine Learning Algorithms for Fake News Detection BT - Advances in Computing and Data Sciences. Springer Singapore.
5. Balyan, R., McCarthy, K. S., & McNamara, D. S. (2020). Applying natural language processing and hierarchical machine learning approaches to text difficulty classification. International Journal of Artificial Intelligence in Education, 30(3), 337–370.
6. Başarslan, M. S., & Kayaalp, F. (2024). Sentiment analysis using a deep ensemble learning model. Multimedia Tools and Applications, 83(14), 42207–42231.
7. Blitzer, J., Dredze, M., & Pereira, F. (2007). Biographies, bollywood, boom-boxes and blenders: Domain adaptation for sentiment classification. Proceedings of the 45th Annual Meeting of the Association of Computational Linguistics, 440–447.
8. Çetiner, H. (2022). Multi-Label Text Analysis with a CNN and LSTM Based Hybrid Deep Learning Model. Journal of Engineering Science of Adiyaman University, 9(17), 15–16.

9. Çetiner, H. (2024a). Fake News Detection and Classification with Recurrent Neural Network Based Deep Learning Approaches. Osmaniye Korkut Ata Üniversitesi Fen Bilimleri Enstitüsü Dergisi, 7(3), 973–993.
10. Çetiner, H. (2024b). Skin Lesions Identification and Analysis with Deep Learning Model Using Transfer Learning. Osmaniye Korkut Ata Üniversitesi Fen Bilimleri Enstitüsü Dergisi, 7(3), 1030–1045.
11. Çetiner, H., & Metlek, S. (2024). Analysis of Different Pooling Functions on A Convolution Neural Network Based Model. International Journal of 3D Printing Technologies and Digital Industry, 8(2), 266–276.
12. Çetiner, M. (2022). Analysis of sustainable fashion products using a deep learning approach (PhD thesis). Süleyman Demirel University, Faculty of Business Administration.
13. Chen, T., Xu, R., He, Y., & Wang, X. (2017). Improving sentiment analysis via sentence type classification using BiLSTM-CRF and CNN. Expert Systems with Applications, 72, 221–230.
14. Cheng, Y., Sun, H., Chen, H., Li, M., Cai, Y., Cai, Z., & Huang, J. (2021). Sentiment Analysis Using Multi-Head Attention Capsules With Multi-Channel CNN and Bidirectional GRU. IEEE Access, 9, 60383–60395.
15. Cheng, Y., Ye, Z. M., Wang, M. W., Zhang, Q., & Zhang, G. H. (2019). Analysis of Chinese text sentiment orientation based on convolutional neural network and hierarchical attention network. The Chinese Journal of Process Engineering, 33(1), 133–142.
16. Cho, K., Van Merriënboer, B., Gulcehre, C., Bahdanau, D., Bougares, F., Schwenk, H., & Bengio, Y. (2014). Learning phrase representations using RNN encoder-decoder for statistical machine translation. ArXiv Preprint ArXiv:1406.1078.
17. Dai, Y., Wu, Z., & Zhang, H. (2021). Sentiment Analysis of Comment Texts Based on CNN-BiGRU-Attention. 2021 China Automation Congress (CAC), 2749–2754.
18. Dang, N. C., Moreno-García, M. N., & De la Prieta, F. (2020). Sentiment Analysis Based on Deep Learning: A Comparative Study. In Electronic, 9(3).
19. Devlin, J. (2018). Bert: Pre-training of deep bidirectional transformers for language understanding. ArXiv Preprint ArXiv:1810.04805.
20. Eight, F. (2019). Twitter Airline Sentiment. URL: https://www.kaggle.com/datasets/crowdflower/twitter-airline-sentiment (accessed date: March 24, 2024).
21. Evaluation, S. (2017). SemEval-2017. URL: https://alt.qcri.org/semeval2017/ (accessed date: March 24, 2024).
22. Go, A., Bhayani, R., & Huang, L. (2009). Twitter sentiment classification using distant supervision. CS224N Project Report, Stanford, 1(12), 2009.
23. Graves, A. (2013). Generating sequences with recurrent neural networks. ArXiv Preprint ArXiv:1308.0850.
24. Karcioğlu, A. A., & Aydin, T. (2019). Sentiment Analysis of Turkish and English Twitter Feeds Using Word2Vec Model. 2019 27th Signal Processing and Communications Applications Conference (SIU), 1–4.
25. Karcioğlu, A. A., & Bulut, H. (2021). Performance Evaluation of Classification Algorithms Using Hyperparameter Optimization. 2021 6th International Conference on Computer Science and Engineering (UBMK), 354–358.
26. Karcioğlu, A. A., & Yaşa, A. C. (2020). Automatic Summary Extraction in Texts Using Genetic Algorithms. 2020 28th Signal Processing and Communications Applications Conference (SIU), 1–4.
27. Karcıoğlu, A. A., Tanışman, S., & Bulut, H. (2021). Türkiye’de COVID-19 Bulaşısının ARIMA Modeli ve LSTM Ağı Kullanılarak Zaman Serisi Tahmini. Avrupa Bilim ve Teknoloji Dergisi, 32, 288–297.
28. Kishwar, A., & Zafar, A. (2023). Fake news detection on Pakistani news using machine learning and deep learning. Expert Systems with Applications, 211, 118558.
29. Li, D., Shi, X., & Dai, M. (2024). A Text Sentiment Classification Method Enhanced by Bi-GRU and Attention Mechanism BT-Proceedings of the 13th International Conference on Computer Engineering and Networks. Springer Nature Singapore.
30. Luong, T., Pham, H., & Manning, C. D. (2015). Effective Approaches to Attention-based Neural Machine Translation. In L. Màrquez, C. Callison-Burch, & J. Su (Eds.), Proceedings of the 2015 Conference on Empirical Methods in Natural Language Processing (pp. 1412–1421). Association for Computational Linguistics.
31. Ma, X. (2016). End-to-end Sequence Labeling via Bi-directional LSTM-CNNs-CRF. ArXiv Preprint ArXiv:1603.01354.
32. Metlek, S., & Çetiner, H. (2024). Inception SH: A New CNN Model Based on Inception Module for Classifying Scene Images. Mühendislik Bilimleri ve Tasarım Dergisi, 12(2), 328–344.
33. Mikolov, T., Chen, K., Corrado, G. ., & Dean, J. (2013). Efficient Estimation of Word Representations in Vector Space. Proceedings of Workshop at ICLR, 2013.
34. Mikolov, T., Sutskever, I., Chen, K., Corrado, G. S., & Dean, J. (2013). Distributed representations of words and phrases and their compositionality. Advances in Neural Information Processing Systems, 3111–3119.
35. Mostafa, L. (2020). Machine learning-based sentiment analysis for analyzing the travelers reviews on Egyptian hotels. The International Conference on Artificial Intelligence and Computer Vision, 405–413.
36. Nasukawa, T., & Yi, J. (2003). Sentiment analysis: Capturing favorability using natural language processing. Proceedings of the 2nd International Conference on Knowledge Capture, 70–77.
37. Peslak, A., Hunsinger, S., & Kruck, S. (2018). Text messaging today: A longitudinal study of variables influencing text messaging from 2009 to 2016. Journal of Information Systems Applied Research, 11(3), 25.
38. Priya, C. S. R., & Deepalakshmi, P. (2023). Sentiment analysis from unstructured hotel reviews data in social network using deep learning techniques. International Journal of Information Technology, 15(7), 3563–3574.
39. Salmony, M. Y., Faridi, A. R., & Masood, F. (2023). Leveraging attention layer in improving deep learning models performance for sentiment analysis. International Journal of Information Technology.
40. Shiau, W.-L., Dwivedi, Y. K., & Lai, H.-H. (2018). Examining the core knowledge on facebook. International Journal of Information Management, 43, 52–63.
41. Vaswani, A., Shazeer, N., Parmar, N., Uszkoreit, J., Jones, L., Gomez, A. N., Kaiser, L., & Polosukhin, I. (2017). Attention Is All You Need.(Nips), 2017. ArXiv Preprint ArXiv:1706.03762, 10, S0140525X16001837.
42. Wang, Y., Liao, W., & Chang, Y. (2018). Gated Recurrent Unit Network-Based Short-Term Photovoltaic Forecasting. In Energies, 11(8).
43. Wankhade, M., Annavarapu, C. S. R., & Abraham, A. (2024). CBMAFM: CNN-BiLSTM Multi-Attention Fusion Mechanism for sentiment classification. Multimedia Tools and Applications, 83(17), 51755–51786.
44. Xiao, Z., Li, X., Wang, L., Yang, Q., Du, J., & Sangaiah, A. K. (2018). Using convolution control block for Chinese sentiment analysis. Journal of Parallel and Distributed Computing, 116, 18–26.
45. Yildirim, G. (2022). A novel grid-based many-objective swarm intelligence approach for sentiment analysis in social media. Neurocomputing, 503, 173–188. https://doi.org/10.1016/j.neucom.2022.06.092.
46. Zhang, M., Zheng, R., Chen, J., Zhu, J., Liu, R., Sun, S., & Wu, Q. (2019). Emotional component analysis and forecast public opinion on micro-blog posts based on maximum entropy model. Cluster Computing, 22, 6295–6304.
47. Zhang, W., Li, X., & Liu, W. (2022). A Review of Text Sentiment Analysis. International Journal of Social Science and Education Research, 5(9), 23–28.
48. Zhao, Z., & Wu, Y. (2016). Attention-Based Convolutional Neural Networks for Sentence Classification. Interspeech, 8, 705–709.
49. Zhou, Y. (2019). Sentiment classification with deep neural networks (Master of Science Thesis). Tampere University, Faculty of Information Technology and Communication Sciences.
50. Zulqarnain, M., Ghazali, R., Aamir, M., & Hassim, Y. M. M. (2024). An efficient two-state GRU based on feature attention mechanism for sentiment analysis. Multimedia Tools and Applications, 83(1), 3085–3110