Design and Implementation of Personalized Recommendation Algorithm in Alumni Using Natural Language Processing

Recently, new possibilities for its web portal application have been presented by the automatic discovery of information in educational data, which has been expanding its horizons. The recommendation of university courses for high school students is an unexplored area. Traditional recommendation systems rely on collaborative filtering, which, in this instance, does not apply since the number of items and users needed to achieve high-performance levels is too small. One of the most well-known recommendation approaches in the course recommendation career is Natural Language Processing (NLP). Association rules have been added using data from alumni and assessments of student recommendations. The rules have been created by first mining data about the course that students completed, then comparing the discovered relationships to the existing course tree to improve it. Hence, a Natural Language Processing-based Personalized Recommendation System (NLP-PRS) has been established to help students choose courses that would take them to their desired careers by analyzing data from alumni and job offers. This study can use the alumni web portal and content formation through NLP to extract relevant information from course programs. It then presents recommendations based on students’ performance, interests, and results in each course that makes up each program. In addition, for the best results in improving their learning skills, students’ decisions should inform the recommended compliance. Because students vary in preferences (based on prior knowledge, learning style, emotional state, etc.), it is impossible to recommend a set timeframe for the recommended sequence. Based on the experimental outcome data, it gives the impression that the system is superior for allowing alumni to engage, share updates, and interact with current students who are alumni.