A Hypergraph-based Resume-Job Matching Model for Recommendation

In recent years, Al-related technologies have triggered the rapid development of online recruitment. However, existing studies mainly focus on unilateral recommendation that based on the features of one side, and rarely considers the preferences of both job seekers and recruiters. To address this challenge, we propose a novel hypergraph-based resume-job matching model, namely HRJM model, which takes into account the preferences of both sides of job seekers and recruiters. Specifically, we construct a resume feature hypergraph and a job feature hypergraph, which represent the job seekers’ resumes and recruiting positions’ requirements, respectively. Among them, the nodes are the features of each attribute, and the hyperedges connect the features, representing the preferences of each resume and job. Then, we designed a concatenate layer to merge the embedding representation of job seekers’ resumes and recruiting jobs, adequately considering the preferences of both sides in the recruiting market. Finally, we designed a classify module to make predictions for our job-person fit problem. Extensive experiments show that our proposed model outperforms the state-of-the-art methods in various evaluation metrics.