A Feature Fusion-based Representation Learning Model for Job Recommendation
Miao He, Yuanyuan Zhu, Nan Lv, Renjie He · 2022 2nd International Conference on Consumer Electronics and Computer Engineering (ICCECE) · 2022
Recently, lots of online recruitment sites emerged. It is difficult for people to find their interested jobs only with keywords retrieving. An intelligent job recommendation system is efficient to overcome the information overload of online recruitment. Many existing works on job recommendations focus on the textual data of jobs and resumes, which ignoring the structured features. In this work, a feature fusion-based representation learning model is proposed to predict the matching scores between job postings and resumes. Specifically, two modules are proposed to get the hidden vectors of textual and structural features, respectively. By concatenating the vectors from embedding modules, a comprehensive representation of the job posting (or resume) is composed. Last, a bilinear module is utilized to get the matching degree between the job and resume. Experiments over real-world data validate our proposed model outperforms baseline models.