<![CDATA[Twitter (X) is one of the most frequently used social media platforms for people to freely express their opinions, including their perceptions of government performance during flood disasters. Among them, the handling of flood disasters in the Jabodetabek region is a highly discussed topic that causes widespread public reaction. This study aims to classify public sentiment using IndoBERT and a hybrid IndoBERT + CNN-LSTM model. A dataset of 3,894 Indonesian-language tweets was collected, pre-processed, and labelled. The sentiment classification was evaluated using 10-fold cross-validation with accuracy, precision, recall, and F1-score as performance metrics. IndoBERT achieved an accuracy of 91.76% and an F1-score of 90.66%, while the IndoBERT + CNN-LSTM model showed better performance with 94.92% accuracy and a 95.41% F1-score. Although this study used raw tweet locations without sentiment labels for geospatial mapping, the results show a significant improvement in sentiment classification from combining semantic and sequential modelling. For future research, the integration of sentiment data into spatial visualization is recommended to provide deeper insights into regional public opinion.]]>
Copyrights © 2025
