Integrasi Sentiment Analysis pada Smart-App Safe-Child untuk Deteksi Dini Risiko Perilaku Agresif Akibat Bullying

Abstract

Bullying among children can negatively affect emotional well-being and contribute to behavioral problems if not identified at an early stage. This study aims to develop the Safe-Child Smart-App as a tool for monitoring students’ emotional conditions by integrating Sentiment Analysis of daily stories. The study employs the ADDIE development model, consisting of Analysis, Design, Development, Implementation, and Evaluation. The application allows students to select one of five daily mood categories—very happy, neutral, sad, very sad, and angry—and write a story describing their experiences. The stories are analyzed using a sentiment lexicon approach and combined with mood data and historical records to determine attention priorities and provide notifications to school counseling teachers. The development results show that Safe-Child provides features for story recording, mood history and visualization, and a counseling teacher dashboard for monitoring students’ emotional conditions and priority levels. User acceptance evaluation involving 20 respondents across eight indicators resulted in an acceptance rate of 90%. These results indicate that the application is well accepted as a supporting tool for monitoring students’ emotional conditions and helping counseling teachers identify students who may require further attention.