Public Sentiment Analysis on the Increase of Value Added Tax in Indonesia Through Tweet-Harvest

Abstract

The government’s policy to increase the Value Added Tax (VAT) rate in 2025 has sparked various public reactions, particularly on the social media platform Twitter. This study aims to analyze public sentiment toward the policy using the Knowledge Discovery in Database (KDD) approach. Data were collected through Tweet-Harvest from January to May 2025 and processed through several stages, including text preprocessing, transformation into numerical representation using the Term Frequency–Inverse Document Frequency (TF-IDF) method, and feature selection with Information Gain. Sentiment classification was conducted using the Support Vector Machine (SVM) algorithm, while parameter tuning (hyperparameter tuning) was performed via Grid Search to optimize model performance. Model evaluation was carried out using accuracy, precision, recall, and F1-Score metrics. The analysis revealed that public opinions were categorized into three sentiment classes: positive, negative, and neutral, with negative sentiment being the most dominant. These findings provide insight into public perception of the VAT increase and can serve as a reference for the government in developing more effective and responsive policy communication strategies.