This study examines taxpayers’ perceptions of income tax policies by analyzing sentiment expressed in social media posts and corroborating findings with government reports and news articles. Employing sentiment analysis via VADER and BERT classifiers, topic modeling using Latent Dirichlet Allocation (LDA), and word frequency analysis with TF-IDF, we investigate prevailing opinions and concerns related to income tax policies. The results reveal that a significant proportion of the online discourse is negative, highlighting issues such as perceived unfairness and excessive tax burdens. Our findings are discussed in light of existing tax compliance theories and behavioral frameworks. The study offers policy implications for tax administrators aiming to improve public trust and enhance compliance through better communication and targeted reforms.