تحلیل احساسات متون انگلیسی با استفاده از مدل‌های ترکیبی یادگیری عمیق

نوع مقاله : مقاله پژوهشی

نویسنده

گروه مهندسی کامپیوتر، واحد یزد، دانشگاه آزاد اسلامی، یزد، ایران

10.22034/csj.2023.184645

چکیده

با افزایش استفاده از اینترنت و شبکه‌های اجتماعی، حجم انبوهی از نظرات کاربران در ارتباط با موضوعات مختلف تولید می‌گردد. در نتیجه به‌کارگیری تکنیک‏های علمی نوین جهت تحلیل این نظرات جهت افزایش رضایت مشتریان ضروری به نظر می‏رسد. تحلیل احساسات نظرات کاربران، به‌عنوان یک راهکار ویژه و مؤثر، به دنبال کشف دانش از این متون جهت رفع چالش قطبیت آنها می‏باشد. در این تحقیق، رویکردی ترکیبی مبتنی بر دو روش یادگیری عمیق RNN-GRU و مبتنی بر تعبیه‌گذاری کلمات جهت تحلیل احساسات نظرات کاربران ارائه گردیده است. جهت بهبود تعیین قطبیت از تعبیه‌گذاری کلمات از پیش آموزش دیده شده Word2vec و GloVe استفاده شده است. نتایج ارزیابی بر روی دو مجموعه ‌داده توییت‌های خطوط هوایی و نظرات فیلم نشان می‌دهد که روش پیشنهادی از نظر دقت در تعیین قطبیت نظرات، بهبود 1% را نسبت به روش‌های ترکیبی دیگر داشته است.

کلیدواژه‌ها


  1. Alshuwaier, F., Areshey, A., & Poon, J. Applications and Enhancement of Document-Based Sentiment Analysis in Deep learning Methods: Systematic Literature Review. Intelligent Systems with Applications, 2022.
  2. O., Chiclanaa. F., Carter. J., Fujita. H., Successes and challenges in developing a hybrid approach to sentiment analysis, Appl. Intell. vol.48, 1176–1188, 2018.
  3. E., Affective computing and sentiment analysis, IEEE Intell. Syst. Vol. 31, 102–107, 2016.
  4. Gupte, A., Joshi, S., Gadgul, P., Kadam, A., & Gupte, A. Comparative study of classification algorithms used in sentiment analysis. International Journal of Computer Science and Information Technologies, 5(5), 6261-6264, 2014.
  5. Devika, M. D., Sunitha, C., & Ganesh, A. Sentiment analysis: a comparative study on different approaches. Computer Science Procedia, 87, 44-49, 2016.
  6. Schmidhuber, J. Deep learning in neural networks: An overview. Neural networks, 61, 85-117, 2015.
  7. B., Liakata .M., Zubiaga. A., and Procter.R., Tdparse: Multi-target-specific sentiment recognition on twitter, The 15th Conference of the European Chapter of the Association for Computational Linguistics, Vol.1, 483–493, 2017.
  8. Wang, Y., Huang, M., Zhao, L., & Zhu, X., Attention-based LSTM for Aspect-level Sentiment Classification, The Conference on Empirical Methods in Natural Language Processing (EMNLP), 606–615, 2016.
  9. Wang, J., Sun, C., Li, S., Wang, J., Si, L., Zhang, M., ... & Zhou, G. Human-like decision making: Document-level aspect sentiment classification via hierarchical reinforcement learning. arXiv preprint arXiv:1910.09260, 2019.
  10. Wu C, Xiong Q, Gao M, Li Q, Yu Y, Wang K. A relative position attention network for aspect-based sentiment analysis. Knowledge and Information Systems. 1-5, 2020.
  11. Shuang K, Yang Q, Loo J, Li R, Gu M. Feature distillation network for aspect-based sentiment analysis. Information Fusion. Vol.61, 13-23, 2020.
  12. Tay, Y., Tuan, L. A., & Hui, S. C., Dyadic Memory Networks for Aspect-based Sentiment Analysis. The ACM Conference on Information and Knowledge Management, 107-116, 2017.
  13. Abdi A, Shamsuddin SM, Hasan S, Piran J. Deep learning-based sentiment classification of evaluative text based on Multi-feature fusion. Information Processing & Management, Vol. 56(4), 1245-59, 2019.
  14. Mercha, E. M., & Benbrahim, H. Machine learning and deep learning for sentiment analysis across languages: A survey. Neurocomputing, 531, 195-216, 2023.
  15. Otter, D. W., Medina, J. R., & Kalita, J. K. A survey of the usages of deep learning for natural language processing. IEEE transactions on neural networks and learning systems, 32(2), 604-624, 2020.
  16. Almeida, F., & Xexéo, G. Word embeddings: A survey. arXiv preprint arXiv:1901.09069, 2019.
  17. Church, K. W. Word2Vec. Natural Language Engineering23(1), 155-162, 2017.
  18. Pennington, J., Socher, R., & Manning, C. D. Glove: Global vectors for word representation, The 2014 conference on empirical methods in natural language processing (EMNLP), 1532-1543, 2014.
  19. Tanaka, H., Shinnou, H., Cao, R., Bai, J., & Ma, W. Document classification by word embeddings of bert. In Computational Linguistics: 16th International Conference of The Pacific Association for Computational Linguistics, 145-154, 2020.
  20. Gaikwad, V., & Haribhakta, Y. Adaptive glove and fasttext model for hindi word embeddings. The 7th ACM IKDD CoDS and 25th COMAD, 175-179, 2020.
  21. Kim Y. Convolutional neural networks for sentence classification. The Conference on Empirical Methods in Natural Language Processing (EMNLP), Doha,1746–1751, 2014.
  22. B. and Carley .K., Parameterized convolutional neural networks for aspect level sentiment classification, The Conference on Empirical Methods in Natural Language Processing, 1091–1096, 2018.
  23. Wu C, Xiong Q, Gao M, Li Q, Yu Y, Wang K. A relative position attention network for aspect-based sentiment analysis. Knowledge and Information Systems. 1-5, 2020.
  24. Hyun D, Park C, Yang MC, Song I, Lee JT, Yu H. Target-aware convolutional neural network for target-level sentiment analysis. Information Sciences. Vol. 491, 166-178, 2019.
  25. Kalchbrenner N, Grefenstette E, Blunsom P. A convolutional neural network for modeling sentences. The 52nd Annual Meeting of the Association for Computational Linguistics, 655–665, 2014.
  26. Yin, R., Li, P., & Wang, B. Sentiment lexical-augmented convolutional neural networks for sentiment analysis. IEEE Second International Conference on Data Science in Cyberspace (DSC), 630-635, 2017.
  27. Guan Z, Chen L, Zhao W, Zheng Y, Tan S, and Cai D. Weakly-supervised deep learning for customer review sentiment classification. The International Joint Conference on Artificial Intelligence (IJCAI), 3719-3725,
  28. Mishra A, Dey K, Bhattacharyya P. Learning cognitive features from gaze data for sentiment and sarcasm classification using convolutional neural network. The 55th Annual Meeting of the Association for Computational Linguistics, 377-387, 2017.
  29. Bian, Y., Ye, R., Zhang, J., & Yan, X. Customer preference identification from hotel online reviews: A neural network based fine-grained sentiment analysis. Computers & Industrial Engineering, 172, 2022.
  30. Zharmagambetov AS, Pak AA. Sentiment analysis of a document using deep learning approach and decision trees. In Twelve international conference on electronics computer and computation (ICECCO), 1-4, 2015.
  31. Huang M, Qian Q, Zhu X. Encoding syntactic knowledge in neural networks for sentiment classification. ACM Transactions on Information Systems (TOIS). 35(3), 1-27, 2017.
  32. Mousa A, Schuller B. Contextual bidirectional long short-term memory recurrent neural network language models: a generative approach to sentiment analysis. The 15th Conference of the European Chapter of the Association for Computational Linguistics, Valencia, 1023–1032, 2017.
  33. Teng Z, Vo DT, Zhang Y. Context-sensitive lexicon features for neural sentiment analysis. The conference on empirical methods in natural language processing, 1629-1638, 2016.
  34. Qian Q, Huang M, Lei J, and Zhu X. Linguistically regularized LSTM for sentiment classification. The Annual Meeting of the Association for Computational Linguistics, 1365-1374, 2017.
  35. Rong W, Nie Y, Ouyang Y, et al. Auto-encoder based bagging architecture for sentiment analysis. J Visual Lang Comput, Vol, 25, 840–849, 2014.
  36. Jin Y, Zhang H, Du D L. Improving deep belief networks via delta rule for sentiment classification. The 28th International Conference on Tools with Artificial Intelligence, 410–414, 2016.
  37. Jin Y, Zhang H, Du D L. Incorporating positional information into deep belief networks for sentiment classification, Industrial Conference on Data Mining, pp.1–15, 2017.
  38. Vlachostergiou A, Caridakis G, Mylonas P, et al. Learning representations of natural language texts with generative adversarial networks at document, sentence, and aspect level. Algorithms, Vol.11: 164, 2018.
  39. Socher R, Pennington J, Huang EH, Ng AY, Manning CD. Semi-supervised recursive autoencoders for predicting sentiment distributions. The conference on empirical methods in natural language processing, 151-161, 2011.
  40. Wang X, Jiang W, Luo Z. Combination of convolutional and recurrent neural network for sentiment analysis of short texts. The 26th international conference on computational linguistics: Technical papers, 2428-2437, 2016.
  41. Guggilla C, Miller T, Gurevych I. CNN-and LSTM-based claim classification in online user comments. The 26th International Conference on Computational Linguistics: Technical Papers, 2740-2751, 2016.
  42. Chen T, Xu R, He Y, Wang X. Improving sentiment analysis via sentence type classification using BiLSTM-CRF and CNN. Expert Systems with Applications. Vol. 72, 221-30, 2017.
  43. Chavosh, N., Emadi, S, Improving Sentence Polarity Determination in Sentiment Analysis based on RNN andLSTM Deep Learning Algorithm, Applied and Basic Machine Intelligence Research, 1(1), 108-115, 2022.
  44. Zhang, B., et al., Sentiment analysis through critic learning for optimizing convolutional neural networks with rules. Neurocomputing, 2019. 356: p. 21-30.
  45. Li, L., Goh, T. T., & Jin, D. (2020). How textual quality of online reviews affect classification performance: a case of deep learning sentiment analysis. Neural Computing and Applications, 32(9), 4387-4415.
  46. Mandhula, T., Pabboju, S., & Gugulotu, N. (2019). Predicting the customer’s opinion on amazon products using selective memory architecture-based convolutional neural network. The Journal of Supercomputing, 1-25.
  47. Wei, J., et al., BiLSTM with multi-polarity orthogonal attention for implicit sentiment analysis. Neurocomputing, 2020. 383: p. 165-173.
  48. Dashtipour, K., et al., A hybrid Persian sentiment analysis framework: Integrating dependency grammar based rules and deep neural networks. Neurocomputing, 2020. 380: p. 1-10.
  49. Das, S., Das, D., & Kolya, A. K. (2020). Sentiment classification with GST tweet data on LSTM based on polarity-popularity model. Sādhanā, 45, 1-17.
  50. Colón-Ruiz, C. and I. Segura-Bedmar, Comparing deep learning architectures for sentiment analysis on drug reviews. Journal of Biomedical Informatics, 2020. 110: p. 103539.
  51. Heikal, M., M. Torki, and N. El-Makky, Sentiment analysis of Arabic tweets using deep learning. Procedia Computer Science, 2018. 142: p. 114-122.
  52. Chiarello, F., A. Bonaccorsi, and G. Fantoni, Technical sentiment analysis. Measuring advantages and drawbacks of new products using social media. Computers in Industry, 2020. 123: p. 103299.
  53. Ray, B., A. Garain, and R. Sarkar, An ensemble-based hotel recommender system using sentiment analysis and aspect categorization of hotel reviews. Applied Soft Computing, 2021. 98: p. 106935.
  54. Song, C., et al., SACPC: A framework based on probabilistic linguistic terms for short text sentiment analysis. Knowledge-Based Systems, 2020. 194: p. 105572.
  55. Hajipour, O., Sadidpour, S. S., Automatic extraction of keywords from Persian short texts using word2vec., Electronic and cyber defense journal, 2020, 8(2), p. 105-114 (in persian).
  56. Wang, Z., & Lin, Z. (2020). Optimal feature selection for learning-based algorithms for sentiment classification. Cognitive Computation, 12(1), 238-248.
  57. Huddar, M. G., Sannakki, S. S., & Rajpurohit, V. S. (2019). Multi-level context extraction and attention-based contextual inter-modal fusion for multimodal sentiment analysis and emotion classification. International Journal of Multimedia Information Retrieval, 1-10.
  58. Saeed, R. M., Rady, S., & Gharib, T. F. (2021). Optimizing Sentiment Classification for Arabic Opinion Texts. Cognitive Computation, 13(1), 164-178.
  59. Sridharan, K., & Komarasamy, G. (2020). Sentiment classification using harmony random forest and harmony gradient boosting machine. Soft Computing, 24(10), 7451-7458.