ANALISIS SENTIMEN DI INSTAGRAM TERHADAP MENTERI KEUANGAN PURBAYA YUDHI SADEWA MENGGUNAKAN METODE LOGISTIC REGRESSION

Abstract

Instagram is one of the social media platforms used by the public to express opinions on government policies and public figures. This study focuses on the sentiment analysis of netizens' comments on the Minister of Finance of the Republic of Indonesia, Purbaya Yudhi Sadewa, who replaced Sri Mulyani Indrawati in September 2025. The research data consisted of 1,277 comments crawled from Instagram. The data processing stage included text preprocessing, labeling using a lexicon-based approach, and feature extraction with Term Frequency – Inverse Document Frequency (TF-IDF). Sentiment classification was performed using the Logistic Regression algorithm, while model performance evaluation used a confusion matrix, classification reports, and the ROC-AUC Curve. The test results showed an accuracy of 91.4% with a macro f1-score of 0.92. In the positive class, a precision of 0.89, a recall of 0.85, and an f1-score of 0.87 were obtained. The negative class obtained a precision of 0.88, a recall of 0.91, and an f1-score of 0.90. The neutral class performed perfectly, with precision, recall, and an f1-score of 1.00 each. The macro-average ROC-AUC value reached 0.985, indicating excellent model performance. The sentiment distribution was predominantly neutral (42.2%), followed by negative (33.2%) and positive (24.6%). This study provides an objective overview of public perception, with neutral opinions predominating, while criticism outweighed appreciation.