Research on the Enhanced DistilBERT Model for NLP-Based Text Analysis

In this study, deep sentiment analysis of social media text data is conducted using DistilBERT, a deep learning model, and the results are compared with those derived from traditional sentiment analysis methods, including the Affective Norms for English Words (AFINN) sentiment dictionary and support vector machine (SVM) models. This research enhances the DistilBERT model by incorporating advanced Transformer technologies and integrating big data analysis techniques to provide a comprehensive assessment of public sentiment. The model achieves an accuracy of 83.76%, with even better overall performance metrics. The findings indicate that public sentiment towards AI technologies is predominantly neutral, accounting for 51.46% of the sentiments analyzed. This study assesses public emotions towards emerging technologies, exploring their spread and impact on acceptance across user groups. Insights into social media emotions guide future research and applications.