Rating Prediction Based Job Recommendation Service for College Students

When college students enter the job market, one of the main difficulties is that they do not have much working experience. To help students find proper jobs, appropriate recommendation systems are becoming a necessity. However, since most students start to find jobs in a very short time, it is difficult for a recommender system due to the lack of history information. To solve this problem, in this research we proposed a rating prediction mechanism by considering the feedback from graduates who have offers and also provided ratings to the employers. By calculating the similarity between the students, a rating prediction method is proposed to generate a list of potential employers for the students. Furthermore, we also take into account the factor of student's interest into the recommendation list's generation to further polish the overall performance. Experimental study on real recruitment dataset has shown the model's potential.