English writing texts often show complex semantic layers, implicit emotional expression, and strong temporal dependence, which leads to limitations in semantic modeling and restricted feature extraction in existing methods. To address this issue, the study constructed a fine-grained sentiment recognition model that integrates bidirectional encoder representations from transformers (BERT), bidirectional gated recurrent unit (BiGRU), convolutional neural network (CNN), and an attention mechanism. BERT was used to generate context-aware semantic representations and improve overall semantic understanding of the text. BiGRU was applied to capture bidirectional temporal dependencies and describe the dynamic evolution of emotions in discourse. CNN was employed to extract phrase-level local emotional features and enhance the detection of emotion-triggering segments. The attention mechanism was introduced to highlight key emotional information and improve feature discriminability. On this basis, a gating fusion strategy was used to dynamically integrate multi-source features, and a multi-task learning framework was incorporated. The model performed emotion intensity prediction while conducting multi-class emotion classification. In this way, fine-grained sentiment modeling was achieved from both category and intensity perspectives. The results indicate that the proposed model achieves excellent performance across several datasets, with overall accuracy and F1-score remaining above 91%, ROC-AUC reaching up to 0.944, and recognition rates for all emotion categories staying above 84%, clearly outperforming existing mainstream approaches. Ablation results demonstrated that all six modules contributed to performance improvement. On the SST-2 and IMDB datasets, the complete model achieved an accuracy of 93% and an AUC of 0.97. In summary, this research offers a compact, stable, and adaptable solution for emotion modeling in English texts, with both theoretical and practical value.

1. [1] B. Wan, P. Wu, C. K. Yeo, and G. Li, “Emotion-cognitive reasoning integrated BERT for sentiment analysis of online public opinions on emergencies,” Inf. Process. Manag., Vol.61, No.2, Article No.103609, 2024. https://doi.org/10.1016/j.ipm.2023.103609
2. [2] I. Üveges and O. Ring, “HunEmBERT: A fine-tuned BERT-model for classifying sentiment and emotion in political communication,” IEEE Access, Vol.11, pp. 60267-60278, 2023. https://doi.org/10.1109/ACCESS.2023.3285536
3. [3] G. Singh, D. Brahma, P. Rai, and A. Modi, “Text-based fine-grained emotion prediction,” IEEE Trans. Affect. Comput., Vol.15, No.2, pp. 405-416, 2023. https://doi.org/10.1109/TAFFC.2023.3298405
4. [4] D. Kim and P. Kang, “Cross-modal distillation with audio–text fusion for fine-grained emotion classification using BERT and Wav2vec 2.0,” Neurocomputing, Vol.506, pp. 168-183, 2022. https://doi.org/10.1016/j.neucom.2022.07.035
5. [5] Y. He, J. Chen, D. Antonyrajah, and I. Horrocks, “BERTMap: A BERT-based ontology alignment system,” Proc. AAAI Conf. Artif. Intell., Vol.36, No.5, pp. 5684-5691, 2022. https://doi.org/10.1609/aaai.v36i5.20510
6. [6] T. Han, Z. Zhang, M. Ren et al., “Text emotion recognition based on XLNet-BiGRU-Att,” Electronics, Vol.12, No.12, Article No.2704, 2023. https://doi.org/10.3390/electronics12122704
7. [7] J. Deng and Y. Liu, “Research on sentiment analysis of online public opinion based on RoBERTa–BiLSTM–attention model,” Appl. Sci., Vol.15, No.4, Article No.2148, 2025. https://doi.org/10.3390/app15042148
8. [8] Y. Xu and J. Wang, “Design and implementation of information system for vocational education quality assessment based on CNN-BiGRU,” Syst. Soft Comput., Vol.7, Article No.200359, 2025. https://doi.org/10.1016/j.sasc.2025.200359
9. [9] J. Alghamdi, Y. Lin, and S. Luo, “Enhancing hierarchical attention networks with CNN and stylistic features for fake news detection,” Expert Syst. Appl., Vol.257, Article No.125024, 2024. https://doi.org/10.1016/j.eswa.2024.125024
10. [10] M. Khadhraoui, H. Bellaaj, M. B. Ammar, H. Hamam, and M. Jmaiel, “Survey of BERT-base models for scientific text classification: COVID-19 case study,” Appl. Sci., Vol.12, No.6, Article No.2891, 2022. https://doi.org/10.3390/app12062891
11. [11] N. M. Gardazi, A. Daud, M. K. Malik et al., “BERT applications in natural language processing: A review,” Artif. Intell. Rev., Vol.58, Article No.166, 2025. https://doi.org/10.1007/s10462-025-11162-5
12. [12] G. Nkhata, S. Gauch, U. Anjum et al., “Fine-tuning BERT with bidirectional LSTM for fine-grained movie reviews sentiment analysis,” arXiv preprint, arXiv:2502.20682, 2025. https://doi.org/10.48550/arXiv.2502.20682
13. [13] K. L. Tan, C. P. Lee, K. S. M. Anbananthen et al., “RoBERTa-LSTM: A hybrid model for sentiment analysis with transformer and recurrent neural network,” IEEE Access, Vol.10, pp. 21517-21525, 2022. https://doi.org/10.1109/ACCESS.2022.3152828
14. [14] J. Chen and L. Pan, “An AI-based cross-language aspect-level sentiment analysis model using English corpus,” Eng. Rep., Vol.6, No.12, Article No.e12969, 2024. https://doi.org/10.1002/eng2.12969
15. [15] J. Zhang, C. Li, G. Liu, M. Min, C. Wang, J. Li, Y. Wang, H. Yan, Z. Zuo, W. Huang, and H. Chen, “A CNN-transformer hybrid approach for decoding visual neural activity into text,” Comput. Methods Programs Biomed., Vol.214, Article No.106586, 2022. https://doi.org/10.1016/j.cmpb.2021.106586
16. [16] F. Yang, “CNN-transformer triple classification of news text based on multiple strategy optimizations,” Procedia Comput. Sci., Vol.266, pp. 891-898, 2025. https://doi.org/10.1016/j.procs.2025.08.111
17. [17] J. Zhang, J. Wu, H. Wang, Y. Wang, and Y. Li, “Cloud detection method using CNN based on cascaded feature attention and channel attention,” IEEE Trans. Geosci. Remote Sens., Vol.60, Article No.4104717, 2021. https://doi.org/10.1109/TGRS.2021.3120752
18. [18] Q. Zhu, X. Jiang, and R. Ye, “Sentiment analysis of review text based on BiGRU-attention and hybrid CNN,” IEEE Access, Vol.9, pp. 149077-149088, 2021. https://doi.org/10.1109/ACCESS.2021.3118537
19. [19] P. Ma, G. Li, H. Zhang, C. Wang, and X. Li, “Prediction of remaining useful life of rolling bearings based on multiscale efficient channel attention CNN and bidirectional GRU,” IEEE Trans. Instrum. Meas., Vol.73, Article No.2508413, 2024. https://doi.org/10.1109/TIM.2023.3347787
20. [20] Z. Jin and K. Jia, “A temporal multi-scale hybrid attention network for sleep stage classification,” Med. Biol. Eng. Comput., Vol.61, pp. 2291-2303, 2023. https://doi.org/10.1007/s11517-023-02808-z
21. [21] L. Na, T. Cao, S. Bai, and D. Li, “A parallel two-channel emotion classification method for Chinese text,” IEEE Access, Vol.12, pp. 39533-39548, 2024. https://doi.org/10.1109/ACCESS.2024.3350190
22. [22] X. Liu, S. Li, and M. Wang, “Hierarchical attention-based multimodal fusion network for video emotion recognition,” Comput. Intell. Neurosci., Vol.2021, No.1, Article No.5585041, 2021. https://doi.org/10.1155/2021/5585041
23. [23] P. Foggia, A. Greco, A. Saggese, and M. Vento, “Multi-task learning on the edge for effective gender, age, ethnicity and emotion recognition,” Eng. Appl. Artif. Intell., Vol.118, Article No.105651, 2023. https://doi.org/10.1016/j.engappai.2022.105651
24. [24] A. Kocak, E. Erdem, and A. Erdem, “A gated fusion network for dynamic saliency prediction,” IEEE Trans. Cogn. Dev. Syst., Vol.14, No.3, pp. 995-1008, 2021. https://doi.org/10.1109/TCDS.2021.3094974
25. [25] T. Zhang, C. Liu, X. Liu, Y. Liu, L. Meng, L. Sun, W. Jiang, F. Zhang, J. Zhao, and Q. Jin, “Multi-task learning framework for emotion recognition in-the-wild,” Eur. Conf. Comput. Vis., pp. 143-156, 2022. https://doi.org/10.1007/978-3-031-25075-0_11
26. [26] Q. Na, X. Li, Y. Wang, J. Li, Y. Yang, and H. Zhang, “A pre-training method inspired by large language model for power named entity recognition,” Proc. 5th Int. Conf. Internet Things, Autom. Artif. Intell., pp. 308-312, 2023. https://doi.org/10.1145/3653081.3653131
27. [27] R. Bensoltane and T. Zaki, “Combining BERT with TCN-BiGRU for enhancing Arabic aspect category detection,” J. Intell. Fuzzy Syst., Vol.44, No.3, pp. 4123-4136, 2023. https://doi.org/10.3233/JIFS-221214
28. [28] A. Kamal, P. Mohankumar, and V. K. Singh, “IMFinE: An integrated BERT-CNN-BiGRU model for mental health detection in financial context on textual data,” Proc. 19th Int. Conf. Nat. Lang. Process. (ICON), pp. 139-148, 2022. https://aclanthology.org/2022.icon-main.19/
29. [29] X. Bi and T. Zhang, “Pedagogical sentiment analysis based on the BERT-CNN-BiGRU-attention model in the context of intercultural communication barriers,” PeerJ Comput. Sci., Vol.10, Article No.e2166, 2024. https://doi.org/10.7717/peerj-cs.2166