A Collaborative Filtering Recommendation Algorithm using User Implicit Demographic Information

With development of network technology and society, people enjoy the e-commerce shopping convenience while also deeply troubled by the "information overload" problem. Recommendation systems help the customers find suitable products they need from a large number of commodities. Among the recommendation systems, the most widely used algorithm is collaborative filtering (CF) recommendation algorithm. In order to improve the recommendation quality, many scholars combined demographic information with CF algorithm, but they did not take into account the user implicit demographic information. However, there is a gap between explicit demographic information and implicit demographic information. To solve this problem, we propose a way to mine the user implicit demographic information .Then we introduce uncertain multiple attribute decision making method into our algorithm to find out a set of initial items. Finally, we recommend items users might like in the set of initial items according to the similarity. Experiments show that this method is more reasonable and more accurate to make recommendations for the target users.