<![CDATA[Preparing a strong portfolio is a crucial aspect for students in entering the workforce, one of which can be achieved through participation in various competitions. However, selecting competitions that align with student competencies remains a challenge due to the abundance of competition information, diversity in student interests and abilities, and limitations in budget, time, and resources. This study develops a recommendation system based on a Hybrid Recommendation System designed to map student competencies to relevant competition types. The system integrates the Naive Bayes method to classify student competencies and Item-Based Collaborative Filtering to calculate similarities between competition types based on other users’ preferences. The system is developed incrementally using the waterfall approach, including the stages of planning, analysis, design, implementation, and testing. The model follows standard machine learning workflows, comprising data collection, exploration and preprocessing, model building, performance evaluation, and method integration. The research data includes student profiles, competencies, and competition preferences collected through surveys and internal databases. Evaluation results indicate that the system successfully provides relevant competition recommendations with an accuracy rate of 70%. These results demonstrate the system’s contribution in assisting students to select competitions that match their competencies, presented in a user-friendly web-based application.]]>
