Comparison Performance of SVM, Naïve Bayes and XGBoost Classifier on Climate Change Issue
Khusman Anhsori, Guruh Fajar Shidik · 2024
A long-term shift in the weather patterns from tropical to polar regions is called climate change. This is a worldwide threat that is beginning to materialize and exerting pressure on numerous industries. One step in reducing the effects of climate change is measuring public opinion. Every day, thousands of people share thoughts via tweets on the well-known microblogging platform namely Twitter. Twitter, now it’s famous as X, is a fantastic resource for information about public opinion and perceived issue risk. One of the hot topics being discussed on Twitter is climate change. Climate change is one of the well-known and quickly expanding topics of study in sentiment analysis within natural language processing (NLP) and text classification. In this study, the Support Vector Machine (SVM), Naïve Bayes (NB), and XGBoost (XGB) classifier algorithms were used to examine attitudes about the climate change issue. Data obtained from Kaggle was divided into 4 sentiments, namely News, Pro, Neutral and Anti. Dataset was divided into 80% as training data and 20% as testing data. With an accuracy of 73.32%, the SVM classifier beat the NB and XGB classifier in the classification. Hyperparameter was used to optimize the accuracy, and the result increased for 3% in SVM method.