Implementation of Recommendation System using Content Based Approach for Course Selection of University Students

Modern higher education institutions focus on flexibility and interdisciplinary courses which changed the dynamics of course planning for their students. Today, course structures include a wide range of electives, allowing students to choose what they want to learn. However, with so much variety comes with complexity of choosing courses. A student can feel uncertain, overwhelmed by information, and exhausted by the process of choosing courses that meet their strengths and career aspirations. In this context, there is an evident need to develop and intelligent recommendation system to support learners in making informed, confident decision about their academic courses. The primary objective of this research is develop a personalized recommendation system for a student which suggest a suitable course based on their academic performance and their preferences. This research presents a student-centric, content-based course recommendation framework, which combines academic performance, learner preferences and advanced ensemble learning to facilitate smart course selection in higher education. To solve the problem of cold-start, the proposed model is not rely on historical data or past data which used only directly available learner inputs. To improve the model performance, bagging and stacking techniques is used on ensemble learning model. The study evaluates the model’s effectiveness using various metrics for classification, including accuracy, precision, recall, F1-score and an evaluation of the confusion matrix. The result reflects smooth and balanced performance on training, validation and testing etc. which is an indicator of the model’s ability to generalize.