SITARS: Leveraging Tagging-Based and Collaborative Filtering Recommendations in Student Interest Prediction

Abstract

In contemporary higher education, universities struggle to effectively understand and address the diverse interests of students, often resulting in low engagement and poor attendance at academic and extracurricular events. The lack of a robust system for analyzing student preferences leads to generic event planning strategies that fail to resonate with students, inundating them with irrelevant information and diminishing participation rates. To address this challenge, this study proposes the Student Interest Tendencies and Recommendation System (SITARS), a web-based, data-driven framework designed to predict student interest tendencies and enhance engagement. SITARS employs a combination of Tagging-Based Recommendation and Item-Based Collaborative Filtering to generate personalized event recommendations tailored to individual student preferences in domains such as sports and the arts. By aligning event offerings with student interests, SITARS aims to optimize event planning, improve resource allocation, and enrich the overall student experience. Additionally, the system provides data-driven insights to university decision-makers, enabling a more strategic and informed approach to student engagement. The proposed system is expected to significantly enhance event participation rates and contribute to the development of a more student-centered academic environment.