{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,6,13]],"date-time":"2026-06-13T04:56:35Z","timestamp":1781326595488,"version":"3.54.1"},"reference-count":23,"publisher":"MDPI AG","issue":"6","license":[{"start":{"date-parts":[[2026,6,10]],"date-time":"2026-06-10T00:00:00Z","timestamp":1781049600000},"content-version":"vor","delay-in-days":0,"URL":"https:\/\/creativecommons.org\/licenses\/by\/4.0\/"}],"funder":[{"name":"Soft Science Research Supported by the Shaanxi Provincial Innovation Capacity Program","award":["2025KG-YBXM-150"],"award-info":[{"award-number":["2025KG-YBXM-150"]}]},{"name":"2025 Research Project on Ideological and Political Work by the Xi\u2019an University of Architecture and Technology Association for Ideological and Political Education","award":["SZ2506"],"award-info":[{"award-number":["SZ2506"]}]}],"content-domain":{"domain":[],"crossmark-restriction":false},"short-container-title":["Information"],"abstract":"<jats:p>With the widespread use of social media and online platforms in the dissemination of public opinion within universities, the multi-class classification of risk-related texts has become a critical component of online public opinion analysis in higher education institutions. Existing multi-class risk classification methods often focus on static semantic representations, making it difficult to effectively capture the emotional evolution within texts and the differences between samples, which in turn affects the accuracy of risk classification. To address this, this paper proposes a multi-class risk classification model for university online public opinion that integrates contextual semantic modeling, emotional evolution detection, and adaptive confidence-based feature fusion. The model employs pre-trained BERT for context encoding and, while preserving high-level semantic information, enhances the model\u2019s adaptability to domain-specific features through a selective unfreezing strategy. First, a Bidirectional Gated Recurrent Unit (BiGRU) is introduced to model the emotional evolution trajectory within text sequences, and an emotional transition intensity metric is constructed by calculating the difference between adjacent hidden states, thereby explicitly capturing the magnitude of emotional changes. Additionally, a convolutional feature branch is designed to capture local emotional patterns, enhancing the model\u2019s ability to perceive local risk cues and fine-grained emotional fluctuations. Finally, the Emotion-Adaptive Feature Mixer (EAFM) is introduced. This module adaptively weights global emotional evolution features and local emotional pattern features based on sample confidence to adjust the contributions of different feature branches in risk classification. Experimental results demonstrate that the proposed model exhibits good convergence characteristics in the university online public opinion scenario represented by the CUOPO dataset and demonstrates strong interpretability through attention visualization and confidence coefficient analysis.<\/jats:p>","DOI":"10.3390\/info17060579","type":"journal-article","created":{"date-parts":[[2026,6,11]],"date-time":"2026-06-11T02:59:31Z","timestamp":1781146771000},"page":"579","update-policy":"https:\/\/doi.org\/10.3390\/mdpi_crossmark_policy","source":"Crossref","is-referenced-by-count":0,"title":["A Multi-Class Classification Model for Text Related to Online Public Opinion Risks in Higher Education Institutions Based on Confidence-Aware Dynamic Fusion"],"prefix":"10.3390","volume":"17","author":[{"given":"Xin","family":"Gu","sequence":"first","affiliation":[{"name":"School of Management, Xi\u2019an University of Architecture and Technology, Xi\u2019an 710055, China"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Chengjun","family":"Wang","sequence":"additional","affiliation":[{"name":"School of Management, Xi\u2019an University of Architecture and Technology, Xi\u2019an 710055, China"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Kai","family":"Wang","sequence":"additional","affiliation":[{"name":"College of Information and Control Engineering, Xi\u2019an University of Architecture and Technology, Xi\u2019an 710055, China"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Xiang","family":"Zhao","sequence":"additional","affiliation":[{"name":"College of Information and Control Engineering, Xi\u2019an University of Architecture and Technology, Xi\u2019an 710055, China"}],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"1968","published-online":{"date-parts":[[2026,6,10]]},"reference":[{"key":"ref_1","doi-asserted-by":"crossref","first-page":"990","DOI":"10.20965\/jaciii.2024.p0990","article-title":"Public Opinion Evolution Law and Sentiment Analysis of Campus Online Public Opinion Events","volume":"28","author":"Xu","year":"2024","journal-title":"J. Adv. Comput. Intell. Intell. Inform."},{"key":"ref_2","doi-asserted-by":"crossref","unstructured":"Hidri, A., Alsaif, S.A., Alahmari, M., AlShehri, E., and Sassi Hidri, M. (2025). Opinion Mining and Analysis Using Hybrid Deep Neural Networks. Technologies, 13.","DOI":"10.3390\/technologies13050175"},{"key":"ref_3","unstructured":"Albladi, A., Uddin, M.K., Islam, M., and Seals, C. (2025). TWSSenti: A Novel Hybrid Framework for Topic-Wise Sentiment Analysis on Social Media Using Transformer Models. arXiv."},{"key":"ref_4","unstructured":"Li, Y., Lu, Z., Tang, F., Lai, S., Hu, M., Zhang, Y., Xue, H., Wu, Z., Razzak, I., and Li, Q. (2025). Rhythm of Opinion: A Hawkes-Graph Framework for Dynamic Propagation Analysis. arXiv."},{"key":"ref_5","doi-asserted-by":"crossref","first-page":"103","DOI":"10.1007\/s13278-025-01529-5","article-title":"Topic modelling and sentiment analysis for public opinion mining of the #BBNaija reality TV show: A critical analysis","volume":"15","author":"Salami","year":"2025","journal-title":"Soc. Netw. Anal. Min."},{"key":"ref_6","doi-asserted-by":"crossref","first-page":"243","DOI":"10.54254\/2755-2721\/2025.PO24522","article-title":"A Comparative Study on Social Media Sentiment Classification Models Based on BERT Fine-tuning and Fusion with Sentiment Lexico","volume":"157","author":"Lin","year":"2025","journal-title":"Appl. Comput. Eng."},{"key":"ref_7","doi-asserted-by":"crossref","unstructured":"Deng, J., and Liu, Y. (2025). Research on sentiment analysis of online public opinion based on RoBERTa\u2013BiLSTM\u2013attention model. Appl. Sci., 15.","DOI":"10.3390\/app15042148"},{"key":"ref_8","unstructured":"Devlin, J., Chang, M.-W., Lee, K., and Toutanova, K. (2019, January 2\u20137). Bert: Pre-training of deep bidirectional transformers for language understanding. Proceedings of the 2019 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies, Minneapolis, MN, USA."},{"key":"ref_9","doi-asserted-by":"crossref","first-page":"55225","DOI":"10.1109\/ACCESS.2024.3389974","article-title":"China University Online Public Opinion Risk Dataset","volume":"12","author":"Wang","year":"2024","journal-title":"IEEE Access"},{"key":"ref_10","doi-asserted-by":"crossref","unstructured":"Wei, J., and Zou, K. (2019, January 3\u20137). EDA: Easy Data Augmentation Techniques for Boosting Performance on Text Classification Tasks. Proceedings of the 2019 Conference on Empirical Methods in Natural Language Processing and the 9th International Joint Conference on Natural Language Processing (EMNLP-IJCNLP), Hong Kong, China.","DOI":"10.18653\/v1\/D19-1670"},{"key":"ref_11","doi-asserted-by":"crossref","unstructured":"Karimi, A., Rossi, L., and Prati, A. (2021). AEDA: An Easier Data Augmentation Technique for Text Classification. Findings of the Association for Computational Linguistics: EMNLP 2021, Association for Computational Linguistics.","DOI":"10.18653\/v1\/2021.findings-emnlp.234"},{"key":"ref_12","doi-asserted-by":"crossref","unstructured":"Sun, Y., Tang, D., Duan, N., Gong, Y., Feng, X., Qin, B., and Jiang, D. (2020, January 7\u201312). Neural semantic parsing in low-resource settings with back-translation and meta-learning. Proceedings of the AAAI Conference on Artificial Intelligence, New York, NY, USA.","DOI":"10.1609\/aaai.v34i05.6427"},{"key":"ref_13","doi-asserted-by":"crossref","first-page":"140","DOI":"10.23919\/JCC.2020.03.012","article-title":"Monitoring and early warning of new cyber-telecom crime platform based on BERT migration learning","volume":"17","author":"Zhou","year":"2020","journal-title":"China Commun."},{"key":"ref_14","doi-asserted-by":"crossref","unstructured":"Cho, K., van Merri\u00ebnboer, B., Gulcehre, C., Bahdanau, D., Bougares, F., Schwenk, H., and Bengio, Y. (2014). Learning phrase representations using RNN encoder-decoder for statistical machine translation. arXiv.","DOI":"10.3115\/v1\/D14-1179"},{"key":"ref_15","doi-asserted-by":"crossref","unstructured":"Martineau, J., and Finin, T. (2009, January 17\u201320). Delta TFIDF: An improved feature space for sentiment analysis. Proceedings of the International AAAI Conference on Web and Social Media, San Jose, CA, USA.","DOI":"10.1609\/icwsm.v3i1.13979"},{"key":"ref_16","doi-asserted-by":"crossref","unstructured":"Rose, S.J., Engel, D., Cramer, N., and E Cowley, W. (2010). Automatic keyword extraction from individual documents. Text Mining: Applications and Theory, Wiley.","DOI":"10.1002\/9780470689646.ch1"},{"key":"ref_17","unstructured":"Mihalcea, R., and Tarau, P. (2004, January 25\u201326). Textrank: Bringing order into text. Proceedings of the 2004 Conference on Empirical Methods in Natural Language Processing, Barcelona, Spain."},{"key":"ref_18","doi-asserted-by":"crossref","first-page":"6618452","DOI":"10.1155\/2023\/6618452","article-title":"Integrating BERT embeddings and BiLSTM for emotion analysis of dialogue","volume":"2023","author":"Gou","year":"2023","journal-title":"Comput. Intell. Neurosci."},{"key":"ref_19","doi-asserted-by":"crossref","first-page":"012019","DOI":"10.1088\/1742-6596\/1746\/1\/012019","article-title":"Research on text classification based on bert-bigru model","volume":"1746","author":"Yu","year":"2021","journal-title":"J. Phys. Conf. Ser."},{"key":"ref_20","first-page":"2943","article-title":"BERT-CNN: A Deep Learning Model for Detecting Emotions from Text","volume":"71","author":"Abas","year":"2022","journal-title":"Comput. Mater. Contin."},{"key":"ref_21","unstructured":"Liu, Y., Ott, M., Goyal, N., Du, J., Joshi, M., Chen, D., Levy, O., Lewis, M., Zettlemoyer, L., and Stoyanov, V. (2019). Roberta: A robustly optimized bert pretraining approach. arXiv."},{"key":"ref_22","unstructured":"Lan, Z., Chen, M., Goodman, S., Gimpel, K., Sharma, P., and Soricut, R. (2020, January 26\u201330). ALBERT: A Lite BERT for Self-supervised Learning of Language Representations. Proceedings of the International Conference on Learning Representations (ICLR), Addis Ababa, Ethiopia."},{"key":"ref_23","unstructured":"Zhang, Z., Han, X., Liu, Z., Jiang, X., Sun, M., and Liu, Q. (August, January 28). ERNIE: Enhanced Language Representation with Informative Entities. Proceedings of the 57th Annual Meeting of the Association for Computational Linguistics, Florence, Italy."}],"container-title":["Information"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/www.mdpi.com\/2078-2489\/17\/6\/579\/pdf","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2026,6,13]],"date-time":"2026-06-13T04:11:48Z","timestamp":1781323908000},"score":1,"resource":{"primary":{"URL":"https:\/\/www.mdpi.com\/2078-2489\/17\/6\/579"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2026,6,10]]},"references-count":23,"journal-issue":{"issue":"6","published-online":{"date-parts":[[2026,6]]}},"alternative-id":["info17060579"],"URL":"https:\/\/doi.org\/10.3390\/info17060579","relation":{},"ISSN":["2078-2489"],"issn-type":[{"value":"2078-2489","type":"electronic"}],"subject":[],"published":{"date-parts":[[2026,6,10]]}}}