Klasifikasi Tingkat Kepuasan Pendengar Radio Gagak Rimang Blora Menggunakan Algoritma Decision Tree

Radio Gagak Rimang is a Local Public Broadcasting Institution (LPPL) owned by the Blora Regency Government that serves the function of disseminating public information, preserving local culture, and providing entertainment for the community. Objectively and systematically measuring listener satisfaction is a fundamental necessity to maintain sustainability and improve broadcast quality. This study applies the Decision Tree algorithm with the Gini Index criterion to classify listener satisfaction levels based on three main attributes: signal quality, content quality, and presenter quality. Data were collected through a survey of 200 respondents in Blora Regency using a Likert scale questionnaire (1-5). The satisfaction target variable was classified into three classes: Fairly Satisfied, Satisfied, and Very Satisfied. The data were divided with an 80:20 ratio for model training and testing. The evaluation results show that the Decision Tree model achieved an accuracy of 90.00% on the test data and an average accuracy of 88.00% on 10-fold cross validation. The presenter quality attribute had the highest contribution to the classification process at 47.13%, followed by signal quality at 32.60%, and content quality at 17.97%. These findings demonstrate that the Decision Tree algorithm is effective as a decision-support tool for evaluating local radio listener satisfaction.