مروری بر روش‌های یادگیری عمیق برای طبقه‌بندی هیجان در متن: پیشرفت‌ها، چالش‌ها و فرصت‌ها

نوع مقاله : مقاله مروری

نویسندگان

دانشکده مهندسی، دانشگاه بجنورد، بجنورد، ایران

چکیده

امروزه افراد در بستر وب احساسات و هیجان خود را با کمک ابزارهای ارتباطی مختلف به اشتراک می‌گذارند، که یکی از رایج‌ترین آنها بیان احساسات در محتوای متنی مانند پست‌های رسانه‌های اجتماعی، نظرات فروشگاه‌های برخط و مرورهای کاربران است. تشخیص هیجان در متن شاخه‌ای از تحلیل احساسات است که هدف آن شناسایی انواع مختلف هیجان نویسنده در متن است. این حوزه علمی به تولیدکنندگان و ارائه‌کنندگان خدمات کمک می‌کند تا از نقاط ضعف و قوت خود آگاه شده و خدمات بهتری را برای مشتریان فراهم آورند. در سال‌های اخیر، تشخیص هیجان در متن به دلیل کاربردهای گسترده‌اش در تجارت، اقتصاد، سیاست، پزشکی، روانشناسی و جامعه‌شناسی به یک زمینه تحقیقاتی جذاب تبدیل شده است. در این مقاله مساله طبقه‌بندی هیجان در متن و روش‌های حل آن با تاکید بر یادگیری عمیق بررسی خواهد شد. همچنین شرح مختصری بر جدیدترین راهکارهای یادگیری عمیق ارائه خواهد شد که در سال‌های اخیر برای طبقه‌بندی هیجان در متن مورد استفاده قرار گرفته‌اند. علاوه بر این، تعدادی مجموعه داده برچسب‌گذاری شده، مهم‌ترین مسائل باز در تشخیص هیجان و جهت‌گیری‌های تحقیقاتی آینده نیز مطرح خواهند شد که می‌تواند راهنمای خوبی برای محققین جدید این حوزه باشد.

کلیدواژه‌ها

موضوعات


عنوان مقاله [English]

A survey on deep learning methods for text-based emotion classification: Advances, challenges, and opportunities

نویسندگان [English]

  • Mahdi Rasouli
  • Vahid Kiani
Department of Engineering, University of Bojnord, Bojnord, Iran
چکیده [English]

Today, people on the web share their feelings and emotions with the help of various communication tools, one of the most common of which is the expression of feelings in textual content, such as social media posts, online store reviews, and user reviews. Emotion detection in text is a branch of sentiment analysis that aims to identify different types of human emotion in the text. This scientific field helps manufacturers and service providers to be aware of their weaknesses and strengths, and to provide better services to customers. In recent years, emotion recognition in text has become an attractive research field due to its wide applications in business, economics, politics, medicine, psychology, and sociology. In this article, the problem of emotion classification in text and its solution methods will be investigated with emphasis on deep learning. Also, a brief description of the latest deep learning solutions that have been used in recent years to classify emotion in text will be discussed. In addition, some labelled datasets, the most important open issues in emotion recognition, and future research directions will also be presented, which can be a good guide for new researchers in this field.

کلیدواژه‌ها [English]

  • Natural language processing؛ Sentiment analysis
  • Emotion detection؛ Emotion classification in text؛ Deep learning
[1] C. Zhang and Y. Lu, “Study on artificial intelligence: The state of the art and future prospects,” J. Ind. Inf. Integr., vol. 23, p. 100224, 2021, doi: 10.1016/j.jii.2021.100224.
[2] D.W. Otter, J.R. Medina, and J.K. Kalita, “A survey of the usages of deep learning for natural language processing,” IEEE Trans. Neural Networks Learn. Syst., vol. 32, no. 2, pp. 604-624, 2021, doi: 10.1109/TNNLS.2020.2979670.
[3] A. Khosravi and H. Abdolhosseini, “Personality in social networks using thematic modelling of user feedback,” Soft Comput. J., vol. 11, no. 2, pp. 50-61, 2023, doi: 10.22052/scj.2023.243197.1006 [In Persian].
[4] F. Pourgholamali, M. Kahani, and E. Asgarian, “Exploiting Big Data Technology for Opinion Mining,” Soft Comput. J., vol. 9, no. 1, pp. 26-39, 2020, doi: 10.22052/scj.2021.111450 [In Persian].
[5] M. Birjali, M. Kasri, and A.B. Hssane, “A comprehensive survey on sentiment analysis: Approaches, challenges and trends,” Knowl. Based Syst., vol. 226, p. 107134, 2021, doi: 10.1016/j.knosys.2021.107134.
[6] F.A. Acheampong, C. Wenyu, and H. Nunoo-Mensah, “Text-based emotion detection: Advances, challenges, and opportunities,” Eng. Rep., vol. 2, no. 7, p. e12189, 2020, doi: 10.1002/eng2.12189.
[7] S. Kusal, S. Patil, K. Kotecha, R. Aluvalu, and V. Varadarajan, “AI based emotion detection for textual big data: Techniques and contribution,” Big Data Cong. Comput., vol. 5, no. 3, p. 43, 2021, doi: 10.3390/bdcc5030043.
[8] B. Kratzwald, S. Ilic, M. Kraus, S. Feuerriegel, and H. Prendinger, “Deep learning for affective computing: Text-based emotion recognition in decision support,” Decis. Support Syst., vol. 115, pp. 24-35, 2018, doi: 10.1016/j.dss.2018.09.002.
[9] A. Dzedzickis, A. Kaklauskas, and V. Bucinskas, “Human emotion recognition: Review of sensors and methods,” Sensors, vol. 20, no. 3, p. 592, 2020, doi: 10.3390/s20030592.
[10] M. Schreiner, T. Fischer, and R. Riedl, “Impact of content characteristics and emotion on behavioral engagement in social media: literature review and research agenda,” Electron. Commer. Res., vol. 21, no. 2, pp. 329-345, 2021, doi: 10.1007/s10660-019-09353-8.
[11] P. Ekman and D. Cordaro, “What is meant by calling emotions basic,” Emotion Rev., vol. 3, no. 4, pp. 364-370, 2011, doi: 10.1177/1754073911410740.
[12] P. Ekman, “An argument for basic emotions,” Cogn. Emotion, vol. 6, no. 3-4, pp. 169-200, 1992, doi: 10.1080/02699939208411068.
[13] R.E. Jack, W. Sun, I. Delis, O.G.B. Garrod, and P.G. Schyns, “Four not six: Revealing culturally common facial expressions of emotion,” J. Exp. Psychol. Gen., vol. 145, no. 6, pp. 708-730, 2016, doi: 10.1037/xge0000162.
[14] S. Zad, M. Heidari, J.H. Jones Jr., and O. Uzuner, “Emotion detection of textual data: An interdisciplinary survey,” in IEEE World AI IoT Congress (AIIoT), Seattle, WA, USA, 2021, pp. 255-261, doi: 10.1109/AIIoT52608.2021.9454192.
[15] R. Plutchik, “In search of the basic emotions,” PsycCRITIQUES, vol. 29, no. 6, pp. 511-513, 1984.
[16] E.C.-C. Kao, C.-C. Liu, T.-H. Yang, C.-T. Hsieh, and V.-W. Soo, “Towards text-based emotion detection a survey and possible improvements,” in Int. Conf. Inf. Manag. Eng., Kuala Lumpur, Malaysia, 2009, pp. 70-74. doi: 10.1109/ICIME.2009.113.
[17] N. Sabri, R. Akhavan, and B. Bahrak, “EmoPars: A collection of 30k emotion-annotated Persian social media texts,” in Proc. Student Res. Worksh. Assoc. RANLP 2021, 2021, pp. 167-173.
[18] B. Karimi, (2022, Apr. 03), Persian tweets emotional dataset, [Online]. Available: https://www.kaggle.com/behdadkarimi/persian-tweets-emotional-dataset.
[19] H. Mirzaee, J. Peymanfard, H.H. Moshtaghin, and H. Zeinali, “ArmanEmo: A Persian dataset for text-based emotion detection,” arXiv, CoRR abs/2207.11808, 2022, doi: 10.48550/arXiv.2207.11808.
[20] N. Sabri, (2022, Jan. 31), Persian-Emotion-Detection, [Online]. Available: https://github.com/nazaninsbr/Persian-Emotion-Detection.
[21] Arman Rayan Sharif, (2023, Jan. 06), ArmanEmo, [Online]. Available: https://github.com/Arman-Rayan-Sharif/arman-text-emotion.
[22] F. Zare Mehrjardi, M. Yazdian-Dehkordi, and A. Latif, “Evaluating classical machine learning and deep-learning methods in sentiment analysis of Persian telegram message,” Soft Comput. J., vol. 11, no. 1, pp. 88-105, 2022, doi: 10.22052/scj.2023.246553.1077 [In Persian].
[23] M. Feizi-Derakhshi, Z. Mottaghinia, and M. Asgari-Chenaghlu, “Persian Text Classification Based on Deep Neural Networks,” Soft Comput. J., vol. 11, no. 1, pp. 120-139, 2022, doi: 10.22052/scj.2023.243182.1010 [In Persian].
[24] T. Mikolov, K. Chen, G. Corrado, and J. Dean, “Efficient estimation of word representations in vector space,” in 1st Int. Conf. Learn. Represent. (ICLR), Scottsdale, Arizona, USA, 2013, Workshop Track Proceedings.
[25] P. Bojanowski, E. Grave, A. Joulin, and T. Mikolov, “Enriching word vectors with subword information,” Trans. Assoc. Computat. Linguistics, vol. 5, pp. 135-146, 2017, doi: 10.1162/tacl_a_00051.
[26] J. Pennington, R. Socher, and C. Manning, “Glove: Global vectors for word representation,” in Proc. Conf. Empir. Methods Nat. Lang. Process. (EMNLP), Doha, Qatar: Association for Computational Linguistics, 2014, pp. 1532-1543. doi: 10.3115/v1/D14-1162.
[27] J. Devlin, M.-W. Chang, K. Lee, and K. Toutanova, “BERT: Pre-training of deep bidirectional transformers for language understanding,” in Proc. Conf. North American Chapter Assoc. Comput. Linguis. Hum. Lang. Technol. (NAACL-HLT), Minneapolis, MN, USA, 2019, pp. 4171-4186, doi: 10.18653/v1/n19-1423.
[28] K. Clark, U. Khandelwal, O. Levy, and C.D. Manning, “What does BERT look at? An analysis of BERT’s attention,” in Proc. ACL Worksh. BlackboxNLP: Analyzing Interp. Neural Netw. NLP, Florence, Italy: Association for Computational Linguistics, 2019, pp. 276-286. doi: 10.18653/v1/W19-4828.
[29] A. Vaswani et al., “Attention is all you need,” arXiv, CoRR abs/1706.03762, 2017, doi: 10.48550/arXiv.1706.03762.
[30] E. Batbaatar, M. Li, and K.H. Ryu, “Semantic-emotion neural network for emotion recognition from text,” IEEE Access, vol. 7, pp. 111866-111878, 2019, doi: 10.1109/ACCESS.2019.2934529.
[31] A. Chatterjee, U. Gupta, M.K. Chinnakotla, R. Srikanth, M. Galley, and P. Agrawal, “Understanding emotions in text using deep learning and big data,” Comput. Hum. Behav., vol. 93, pp. 309-317, 2019, doi: 10.1016/j.chb.2018.12.029.
[32] T.T. Sasidhar, P.B, and S.K. P, “Emotion detection in Hinglish (Hindi+English) code-mixed social media text,” Proc. Comput. Sci., vol. 171, pp. 1346-1352, 2020, doi: 10.1016/j.procs.2020.04.144.
[33] M. Abdullah, M. Hadzikadicy, and S. Shaikhz, “SEDAT: Sentiment and emotion detection in Arabic text using CNN-LSTM deep learning,” in 17th IEEE Int. Conf. Mach. Learn. Appl. (ICMLA), Orlando, FL: IEEE, 2018, pp. 835-840, doi: 10.1109/ICMLA.2018.00134.
[34] X. Sun, C. Zhang, and L. Li, “Dynamic emotion modelling and anomaly detection in conversation based on emotional transition tensor,” Inf. Fusion, vol. 46, pp. 11-22, 2019, doi: 10.1016/j.inffus.2018.04.001.
[35] M. Baali and N. Ghneim, “Emotion analysis of Arabic tweets using deep learning approach,” J. Big Data, vol. 6, p. 89, 2019, doi: 10.1186/s40537-019-0252-x.
[36] W. Ragheb, J. Aze, S. Bringay, and M. Servajean, “Attention-based modeling for emotion detection and classification in textual conversations,” arXiv, CoRR abs/1906.07020, 2019.
[37] K. Shrivastava, S. Kumar, and D.K. Jain, “An effective approach for emotion detection in multimedia text data using sequence based convolutional neural network,” Multim. Tools Appl., vol. 78, no. 20, pp. 29607-29639, 2019, doi: 10.1007/s11042-019-07813-9.
[38] A.F. Adoma, N.-M. Henry, W. Chen, and N. Rubungo Andre, “Recognizing emotions from texts using a bert-based approach,” in 17th Int. Comput. Conf. Wavelet Active Media Technol. Inf. Process. (ICCWAMTIP), Chengdu, China: IEEE, 2020, pp. 62-66, doi: 10.1109/ICCWAMTIP51612.2020.9317523.
[39] H. Al-Omari, M.A. Abdullah, and S. Shaikh, “EmoDet2: Emotion detection in English textual dialogue using BERT and BiLSTM models,” in 11th Int. Conf. Inf. Commun. Syst. (ICICS), Irbid, Jordan: IEEE, 2020, pp. 226-232. doi: 10.1109/ICICS49469.2020.239539.
[40] S. Qu, Y. Yang, and Q. Que, “Emotion classification for spanish with XLM-RoBERTa and TextCNN,” in CEUR Workshop Proceedings, Malaga, Spain, September, 2021, pp. 94-100.
[41] A. Das, O. Sharif, M.M. Hoque, and I.H. Sarker, “Emotion classification in a resource constrained language using transformer-based approach,” in Proc. Conf. North American Chapter Assoc. Comput. Linguis. Student Res. Worksh. (NAACL-HLT), Online, 2021, pp. 150-158, doi: 10.18653/v1/2021.naacl-srw.19.
[42] K. Dheeraj and T. Ramakrishnudu, “Negative emotions detection on online mental-health related patients texts using the deep learning with MHA-BCNN model,” Expert Syst. Appl., vol. 182, p. 115265, 2021, doi: 10.1016/j.eswa.2021.115265.
[43] Y. Fu, L. Guo, L. Wang, Z. Liu, J. Liu, and J. Dang, “A sentiment similarity-oriented attention model with multi-task learning for text-based emotion recognition,” in MultiMedia Modeling - Proc. 27th Int. Conf. (MMM), Prague, Czech Republic, 2021, Part I, pp. 278–289, doi: 10.1007/978-3-030-67832-6_23.
[44] X.-M. Lin, C.-H. Ho, L.-T. Xia, and R.-Y. Zhao, “Sentiment analysis of low-carbon travel APP user comments based on deep learning,” Sustain. Energy Technol. Assess., vol. 44, p. 101014, 2021, doi: 10.1016/j.seta.2021.101014.
[45] H. Luo, “Emotion detection for spanish with data augmentation and transformer-based models,” in CEUR Workshop Proceedings, Malaga, Spain, September, 2021, pp. 35-42.
[46] H.Q. Abonizio, E.C. Paraiso, and S. Barbon Junior, “Toward text data augmentation for sentiment analysis,” IEEE Trans. Artif. Intell., vol. 3, no. 5, pp. 657-668, 2022, doi: 10.1109/TAI.2021.3114390.
[47] J. Luo, M. Bouazizi, and T. Ohtsuki, “Data augmentation for sentiment analysis using sentence compression-based SeqGAN with data screening,” IEEE Access, vol. 9, pp. 99922-99931, 2021, doi: 10.1109/ACCESS.2021.3094023.
[48] P. Kumar and B. Raman, “A BERT based dual-channel explainable text emotion recognition system,” Neural Networks, vol. 150, pp. 392-407, 2022, doi: 10.1016/j.neunet.2022.03.017.
[49] A. Majeed, M.O. Beg, U. Arshad, and H. Mujtaba, “Deep-EmoRU: mining emotions from roman urdu text using deep learning ensemble,” Multim. Tools Appl., vol. 81, no. 30, pp. 43163-43188, 2022, doi: 10.1007/s11042-022-13147-w.
[50] X. Zhu, Y. Lou, H. Deng, and D. Ji, “Leveraging bilingual-view parallel translation for code-switched emotion detection with adversarial dual-channel encoder,” Knowl. Based Syst., vol. 235, p. 107436, 2022, doi: 10.1016/j.knosys.2021.107436.
[51] A.B. Warriner, V. Kuperman, and M. Brysbaert, “Norms of valence, arousal, and dominance for 13,915 English lemmas,” Behav. Res., vol. 45, pp. 1191-1207, 2013, doi: 10.3758/s13428-012-0314-x.
[52] M. Lewis et al., “BART: Denoising sequence-to-sequence pre-training for natural language generation, translation, and comprehension,” in Proc. 58th Annu. Meet. Assoc. Comput. Linguist. (ACL), Online, 2020, pp. 7871-7880, doi: 10.18653/v1/2020.acl-main.703.
[53] X. Wu, S. Lv, L. Zang, J. Han, and S. Hu, “Conditional BERT contextual augmentation,” in Computat. Sci. (ICCS), 19th Int. Conf., Faro, Portugal, 2019, pp. 84-95, doi: 10.1007/978-3-030-22747-0_7.
[54] C. Shorten, T.M. Khoshgoftaar, and B. Furht, “Text Data Augmentation for Deep Learning,” J. Big Data, vol. 8, no. 1, p. 101, 2021, doi: 10.1186/s40537-021-00492-0.
[55] S.Y. Feng et al., “A survey of data augmentation approaches for NLP,” in Find. Assoc. Comput. Linguist. (ACL-IJCNLP), Online, 2021, pp. 968-988. doi: 10.18653/v1/2021.findings-acl.84.
[56] A. Khosravi, M. Kelarestaghi, and M. Purmohammad, “Emotion detection in persian text: A machine learning model,” Biannu. J. Iranian Psychol. Assoc., vol. 14, no. 1, pp. 42-48, 2019, doi: 10.29252/bjcp.14.1.42 [In Persian].
[57] S.S. Sadeghi, H. Khotanlou, and M. Rasekh Mahand, “Automatic persian text emotion detection using cognitive linguistic and deep learning,” J. AI Data Mining, vol. 9, no. 2, pp. 169-179, 2021, doi: 10.22044/jadm.2020.9992.2136.
[58] A. Abaskohi, N. Sabri, and B. Bahrak, “Persian emotion detection using ParsBERT and imbalanced data handling approaches,” arXiv, CoRR abs/2211.08029, 2022, doi: 10.48550/arXiv.2211.08029.
[59] A. Khodaei, A. Bastanfard, H. Saboohi, and H. Aligholizadeh, “Deep emotion detection sentiment analysis of persian literary texts,” Research Square preprint, 2022, doi: 10.21203/rs.3.rs-1796157/v1.