An Electronic Commerce Recommendation Approach Based on Time Weight

It has been known that personalized recommendation system is a very important and necessary topic in electronic commerce. Many famous electronic commerce websites employ recommendation systems to convert browsers into buyers. The forms of recommendation include suggesting items to the users, providing personalized service information, summarizing community opinion, and providing society critiques. Collaborative filtering is the most successful technology for building electronic commerce personalized recommendation system and is extensively used in many fields. But traditional collaborative filtering recommendation algorithm does not consider finding the nearest neighbors in different time periods, leading to the neighbors may not be the nearest ones. To solve this issue, an electronic commerce recommendation approach based on time weight is presented. In this method, the time weighted rating is used to search the recommendation items for target users.