Analisis Sentimen Masyarakat Terkait Gaji Pns Rendah Sebagai Faktor Korupsi Memakai Metode Naïve Bayes Classifier(Studi Kasus: Akun Kumparancom Di Instagram)

Abstract

Low salaries of Indonesian civil servants (PNS) are often associated with corruption and widely discussed on social media. This study aims to analyze public sentiment regarding this issue using the Naïve Bayes Classifier method. A total of 1,165 comments were collected from the Kumparancom Instagram account between October 2023 and February 2024 through a crawling process. The data were preprocessed using cleaning, case folding, tokenizing, normalization, stopword removal, and stemming, then classified into positive and negative sentiments. The results indicate that public sentiment is predominantly negative toward the assumption that low civil servant salaries are the main cause of corruption. Model evaluation using a confusion matrix shows that the Naïve Bayes Classifier achieved the highest accuracy of 95% with a 70% training and 30% testing data split. These findings confirm that the proposed method is effective for sentiment analysis of social media text.