Research on the Emotional Classification Model of Online Public Opinion Based on Complex Contexts

Qinyu Ren, Zhiming Zhang, Xiaohua Qiu
Proceedings of 2025 2nd International Conference on Machine Learning and Intelligent Computing, PMLR 278:100-110, 2025.

Abstract

Under the backdrop of the information explosion and the accelerated pace of globalization, online public opinion has become an important channel for reflecting social dynamics and public attitudes. With its huge volume of information, rapid dissemination, diverse viewpoints and ambiguous correlations, it is difficult for traditional sentiment analysis methods to cope with it. This paper aims to construct a sentiment analysis model for online public opinion that can adapt to complex contexts. By means of the Scrapy crawler framework, the public opinion data about the “problematic vaccines" on Sina Weibo is collected. A convolutional neural network situational awareness classification model that combines spatial features and word vectors is proposed. Firstly, preprocessing is carried out based on the spatial distribution characteristics of words in the text. Then, word vectors are constructed based on sentiment features. Finally, the model is constructed and trained. Through comparison with models such as Linear SVM and CNN+Skip-gram, the results show that this model has a certain improvement in both accuracy and recall rate. This paper provides more effective decision-making support and theoretical reference for the analysis of online public opinion, and realizes the improvement of the sentiment classification model in complex contexts.

Cite this Paper

BibTeX
@InProceedings{pmlr-v278-ren25a, title = {Research on the Emotional Classification Model of Online Public Opinion Based on Complex Contexts}, author = {Ren, Qinyu and Zhang, Zhiming and Qiu, Xiaohua}, booktitle = {Proceedings of 2025 2nd International Conference on Machine Learning and Intelligent Computing}, pages = {100--110}, year = {2025}, editor = {Zeng, Nianyin and Pachori, Ram Bilas and Wang, Dongshu}, volume = {278}, series = {Proceedings of Machine Learning Research}, month = {25--27 Apr}, publisher = {PMLR}, pdf = {https://raw.githubusercontent.com/mlresearch/v278/main/assets/ren25a/ren25a.pdf}, url = {https://proceedings.mlr.press/v278/ren25a.html}, abstract = {Under the backdrop of the information explosion and the accelerated pace of globalization, online public opinion has become an important channel for reflecting social dynamics and public attitudes. With its huge volume of information, rapid dissemination, diverse viewpoints and ambiguous correlations, it is difficult for traditional sentiment analysis methods to cope with it. This paper aims to construct a sentiment analysis model for online public opinion that can adapt to complex contexts. By means of the Scrapy crawler framework, the public opinion data about the “problematic vaccines" on Sina Weibo is collected. A convolutional neural network situational awareness classification model that combines spatial features and word vectors is proposed. Firstly, preprocessing is carried out based on the spatial distribution characteristics of words in the text. Then, word vectors are constructed based on sentiment features. Finally, the model is constructed and trained. Through comparison with models such as Linear SVM and CNN+Skip-gram, the results show that this model has a certain improvement in both accuracy and recall rate. This paper provides more effective decision-making support and theoretical reference for the analysis of online public opinion, and realizes the improvement of the sentiment classification model in complex contexts.} }
Endnote
%0 Conference Paper %T Research on the Emotional Classification Model of Online Public Opinion Based on Complex Contexts %A Qinyu Ren %A Zhiming Zhang %A Xiaohua Qiu %B Proceedings of 2025 2nd International Conference on Machine Learning and Intelligent Computing %C Proceedings of Machine Learning Research %D 2025 %E Nianyin Zeng %E Ram Bilas Pachori %E Dongshu Wang %F pmlr-v278-ren25a %I PMLR %P 100--110 %U https://proceedings.mlr.press/v278/ren25a.html %V 278 %X Under the backdrop of the information explosion and the accelerated pace of globalization, online public opinion has become an important channel for reflecting social dynamics and public attitudes. With its huge volume of information, rapid dissemination, diverse viewpoints and ambiguous correlations, it is difficult for traditional sentiment analysis methods to cope with it. This paper aims to construct a sentiment analysis model for online public opinion that can adapt to complex contexts. By means of the Scrapy crawler framework, the public opinion data about the “problematic vaccines" on Sina Weibo is collected. A convolutional neural network situational awareness classification model that combines spatial features and word vectors is proposed. Firstly, preprocessing is carried out based on the spatial distribution characteristics of words in the text. Then, word vectors are constructed based on sentiment features. Finally, the model is constructed and trained. Through comparison with models such as Linear SVM and CNN+Skip-gram, the results show that this model has a certain improvement in both accuracy and recall rate. This paper provides more effective decision-making support and theoretical reference for the analysis of online public opinion, and realizes the improvement of the sentiment classification model in complex contexts.
APA
Ren, Q., Zhang, Z. & Qiu, X.. (2025). Research on the Emotional Classification Model of Online Public Opinion Based on Complex Contexts. Proceedings of 2025 2nd International Conference on Machine Learning and Intelligent Computing, in Proceedings of Machine Learning Research 278:100-110 Available from https://proceedings.mlr.press/v278/ren25a.html.

Related Material