Aspects-Based Sentiment Analysis of Extreme Weather on Twitter Using Long Short-Term Memory

This study presents an aspect-based sentiment analysis of tweets related to extreme weather events in Indonesia, utilizing the Long Short-Term Memory (LSTM) model. The dataset was obtained through a Twitter crawling process, followed by a series of preprocessing steps including data cleaning, stop word removal, normalization, tokenization, and stemming. The three primary areas of emphasis in the study were kinds of bad weather forecasts, and the government or society reactions. Using a lexicon-based technique, sentiment labelling generated three groups: positive, neutral, and negative. A random oversampling method was employed to address the data imbalance. The model using the LSTM algorithm was trained individually for aspect and sentiment classification tasks, so reaching high accuracies of 98.94% and 97.53%, respectively. The results indicate that the model effectively categorises talk on extreme weather and the opinions of the public. A word cloud visual representation was additionally created to show frequently occurring terms in the dataset, thereby offering insights into current themes and sentiment expressions. This work provides valuable input for government agencies and legislators in developing communication and disaster response plans, thereby serving to better understand the public's view on climate-related events. Future work could involve improving techniques for preprocessing and using larger, wider-ranging datasets for improving the model's robustness and generalisation.