Keywords
How to Cite
Abstract
Annotatsiya. Maqolada dialog (suhbat) tizimlarida foydalanuvchi bildirgan hissiyotni aniqlash masalasi uchun diqqat mexanizmiga asoslangan ikki yo'nalishli uzoq qisqa muddatli xotira (BE-Att-BiLSTM) modeli ko'rib chiqiladi. Model uch komponentdan tashkil topgan: oldindan o'qitilgan BERT modelidan olingan kontekstli embedding qatlami, ketma-ketlikning uzoq masofali bog'liqliklarini qamrab oluvchi BiLSTM qatlami hamda hissiyot bashoratiga muhim tokenlarni ajratib beruvchi diqqat qatlami. Model sifatini yanada oshirish maqsadida kontekstli embedding fazosida so'z almashtirishga asoslangan matnli ma'lumotlarni ko'paytirish usuli qo'llanilgan. MELD benchmark to'plamining matnli modalligida o'tkazilgan eksperimentlar diqqat qatlami va ko'paytirish bosqichi qo'shilgan modelning aniqligi 68,00 foizga, F1-balli 67,50 foizga yetishini ko'rsatdi. Olingan natijalar dialog tizimlarida hissiyot tahlili sifatini belgilangan ma'lumotlar cheklangan sharoitda oshirishning amaliy yo'lini asoslaydi.
References
1. Jbene M., Raif M., Tigani S., Chehri A., Saadane R. User sentiment analysis in conversational systems based on augmentation and attention-based BiLSTM // Procedia Computer Science. – 2022. – Vol. 207. – P. 4106–4112. DOI: 10.1016/j.procs.2022.09.473.
2. Ma Y., Nguyen K.L., Xing F.Z., Cambria E. A survey on empathetic dialogue systems // Information Fusion. – 2020. – Vol. 64. – P. 50–70.
3. Wankhade M., Rao A.C.S., Kulkarni C. A survey on sentiment analysis methods, applications, and challenges // Artificial Intelligence Review. – 2022.
4. Al Amrani Y., Lazaar M., El Kadiri K.E. Random forest and support vector machine based hybrid approach to sentiment analysis // Procedia Computer Science. – 2018. – Vol. 127. – P. 511–520.
5. Cheng Y., Yao L., Xiang G. et al. Text sentiment orientation analysis based on multi-channel CNN and bidirectional GRU with attention mechanism // IEEE Access. – 2020. – Vol. 8. – P. 134964–134975.
6. Ghosal D., Majumder N., Gelbukh A.F., Mihalcea R., Poria S. COSMIC: Commonsense knowledge for emotion identification in conversations // Findings of EMNLP 2020. – 2020. – P. 2470–2481.
7. Majumder N., Poria S., Hazarika D. et al. DialogueRNN: An attentive RNN for emotion detection in conversations // Proc. of the 33rd AAAI Conference on Artificial Intelligence. – Honolulu, 2019. – P. 6818–6825.
8. Qin L., Che W., Li Y., Ni M., Liu T. DCR-Net: A deep co-interactive relation network for joint dialog act recognition and sentiment classification // Proc. of the 34th AAAI Conference on Artificial Intelligence. – New York, 2020. – P. 8665–8672.
9. Shorten C., Khoshgoftaar T.M. A survey on image data augmentation for deep learning // Journal of Big Data. – 2019. – Vol. 6, Article 60.
10. Abonizio H.Q., Paraiso E.C., Barbon Junior S. Toward text data augmentation for sentiment analysis // IEEE Transactions on Artificial Intelligence. – 2021. DOI: 10.1109/TAI.2021.3114390.
11. Wei J., Zou K. EDA: Easy data augmentation techniques for boosting performance on text classification tasks // Proc. of EMNLP-IJCNLP 2019. – Hong Kong, 2019. – P. 6381–6387.
12. Liu S., Lee K., Lee I. Document-level multi-topic sentiment classification of email data with BiLSTM and data augmentation // Knowledge-Based Systems. – 2020. – Vol. 197, Article 105918.
13. Devlin J., Chang M.-W., Lee K., Toutanova K. BERT: Pre-training of deep bidirectional transformers for language understanding // Proc. of NAACL-HLT 2019. – Minneapolis, 2019. – P. 4171–4186.
14. Mikolov T., Sutskever I., Chen K., Corrado G., Dean J. Distributed representations of words and phrases and their compositionality // Advances in Neural Information Processing Systems 26. – Lake Tahoe, 2013. – P. 3111–3119.
15. Pennington J., Socher R., Manning C. GloVe: Global vectors for word representation // Proc. of EMNLP 2014. – Doha, 2014. – P. 1532–1543.
16. Schuster M., Paliwal K.K. Bidirectional recurrent neural networks // IEEE Transactions on Signal Processing. – 1997. – Vol. 45, No. 11. – P. 2673–2681.
17. Hochreiter S., Schmidhuber J. Long short-term memory // Neural Computation. – 1997. – Vol. 9, No. 8. – P. 1735–1780.
18. Poria S., Hazarika D., Majumder N., Naik G., Cambria E., Mihalcea R. MELD: A multimodal multi-party dataset for emotion recognition in conversations // Proc. of the 57th Annual Meeting of the ACL. – Florence, 2019. – P. 527–536.