Hybrid Filtering for Student Major Recommendation: A Comparative Study

Choosing the right university major is an important decision for students, as delays or incorrect choices can harm their future careers and cause problems for academic departments. High dropout rates, which are frequently the result of poorly informed decisions, can be a considerable burden on faculty. This project aims to address these challenges by creating a recommendation system that provides individualized counsel to students based on their psychological profiles. A quantitative method was used, with questionnaires distributed to a large number of students. To verify the data's authenticity, replies were sought from students who were pleased with their selected majors rather than those who regretted their choices. The collected data formed the basis for a hybrid recommendation system that integrated Content-based Filtering and Collaborative Filtering methods. The system was then compared against standalone implementations of each filtering method to determine its usefulness in increasing suggestion accuracy. The results showed that the Hybrid Filtering strategy obtained a recommendation accuracy of 84.29%, outperforming Content-based  Filtering at 81.43% and Collaborative Filtering at 78.57%. The proposed model is easy to implement in a school or a university, as long as the required data is available. Thus, the model can help a school or university to reduce dropout rates and boost academic outcomes.