<![CDATA[This study aims to analyze the sentiment of customer complaints regarding Internet Service Provider (ISP) services in Indonesia, where the primary issues frequently reported include connection disruptions, slow internet speeds, weak signals, and unresponsive or uninformative complaint handling, as reflected in various consumer reports on social media. These issues contribute to customer dissatisfaction and necessitate data analysis solutions to deeply understand public opinions. Data was collected via API from a social media platform using keywords related to internet services, such as "internet disruption" and "internet complaints." The data underwent text preprocessing stages, including cleaning, case folding, tokenization, stopword removal, and stemming to produce consistent text. Text features were extracted using Term Frequency–Inverse Document Frequency (TF-IDF), which were then classified using the Support Vector Machine (SVM) algorithm. Model evaluation using 10-Fold Cross Validation yielded an average accuracy of 91.47%, precision of 94.27%, recall of 99.20%, and F1-score of 96.67%. Word frequency analysis revealed dominant words such as “slow,” “disruption,” and “signal” as the main issues in customer complaints. The combination of SVM and TF-IDF proved effective for sentiment analysis in Indonesian, providing academic and practical contributions for ISPs to monitor customer opinions and improve service quality. Future research is recommended to employ deep learning models like BERT and more diverse data.]]>
