doi: 10.7763/IJET.2026.V18.1372
Exploring Public Attitudes toward Artificial Intelligence through Social Media Text Analysis
Xinye Zhou
- Business School, Geely University, Chengdu, China
- Manuscript received September 7, 2026
- accepted October 20, 2026
- published November 12, 2026
Abstract
Sentiment analysis of these texts helps identify the public’s diverse attitudes toward AI. This study focuses on AI-related texts from social media. After text cleaning, manual annotation, and dataset partitioning, a three-class dataset (positive, neutral, and negative) was constructed. Text features were extracted using methods such as TF-IDF and BERT, and classification experiments were conducted using Logistic Regression and SVM. The classification performance of different models was compared using 424 test texts. The results show that the accuracy of the TF-IDF+Logistic Regression, TF-IDF+SVM, BERT+SVM, and BERT+Logistic Regression models were 78.77%, 81.84%, 85.14%, and 86.56%, respectively. Among these, the BERT+Logistic Regression model achieved the best overall classification results, with a Macro-F1 score of 86.32%. Further analysis revealed that the classification performance for neutral text was lower than that for positive and negative text, with misclassifications primarily occurring between neutral text and other categories. The study demonstrates that incorporating contextual semantic features can enhance the ability to identify attitudes in social media text and also highlights the complexity of public attitudes toward artificial intelligence.
Keywords
- social media
- artificial intelligence
- sentiment analysis
- BERT
- text classification
How to Cite
Xinye Zhou, "Exploring Public Attitudes toward Artificial Intelligence through Social Media Text Analysis," International Journal of Engineering and Technology, vol. 18, no. 4, pp. 199-202, 2026. https://doi.org/ 10.7763/IJET.2026.V18.1372
Copyright & License
Copyright © 2026 by the authors. This is an open access article distributed under the Creative Commons Attribution License which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited (CC BY 4.0).