A Deep Hybrid Model for Recommendation Systems

Recommendation has been a long-standing problem in many areas ranging from e-commerce to social websites. Most current studies focus only on traditional approaches such as content-based or collaborative filtering while there are relatively fewer studies in hybrid recommendation systems. With the emergence of deep learning techniques in different fields including computer vision and natural language processing, Recommendation Systems (RSs) have also become an active area of for these techniques. There are several studies that utilize ID embeddings of users and items to implement collaborative filtering with deep neural networks. However, such studies do not take advantage of other categorical or continuous features of inputs. In this paper, we propose a new deep neural network architecture which uses ID embeddings, and also auxiliary information such as features of job postings and candidates. Experimental results on a real world dataset from a job website show that the proposed method improves recommendation results over deep learning models utilizing only ID embeddings.