Prediction of Student Participation in the Library of the University of Muhammadiyah Malang Based on Social Media Activities Using Decision Tree

Abstract

Background: Social media has become an effective tool in promoting library services. This study uses the Decision Tree algorithm to predict student participation in the University of Muhammadiyah Malang library based on their social media activities.Objective: The data used included features such as the type of social media (Instagram, TikTok, YouTube, Twitter), the frequency of visits to the library, and the level of social media engagement.Methods: The decision tree model is processed using Orange Data Mining, which results in a clear separation between participation and non-participation based on a combination of these features.Results: The study results show that social media, especially Instagram and TikTok, significantly influence student participation in the library. The accuracy of the obtained model is about 76.7%, indicating that decision trees are an effective method for predicting library participation.Conclusion: This research provides valuable insights for designing strategies to increase library student engagement based on social media analysis