An Explainable Content-Based Course Recommender Using Job Skills

Abstract

The large number of courses offered in universities and online studies made it difficult for students to choose the courses that suit their interests and career goals, which led students to lose many opportunities to be employed in the job they wanted. To keep pace with the rapid development of technology, and instead of relying on the job title as was previously done, the employers began to identify the skills required for a job. The competencies of the candidates are then examined and evaluated according to those requirements. Thus, it has become necessary for students to take courses that suit their future professional interests, ensuring that they are employed in the job they desire and supporting their long-term career success. Fortunately, the emergence of skills-based employment has provided an opportunity for universities and colleges to create a clearer path to the courses offered to allow students to take courses that match their future career interests. In this study, we used K-Mean clustering algorithm, TF-idf approach, and content-based filtering algorithm to provide relevant courses for students based on the required job with an explanation of why these courses are recommended. Our result illustrates that our method offers many advantages compared with other recommender systems. our system converts a simple course recommendation into a tool for discovering skills.  Since many recommendation systems work as black boxes, we designed our system to recommend the relevant course with explaining why these courses are recommended, which will add a factor of transparency to our system and confirms the reliability of the system to the students.