Research on Application of Collaborative Filtering in Electronic Commerce Recommender Systems

Abstract

In electronic commerce era, personalized recommender systems are popularly being employed to help users in selecting suitable items to meet their personal requirements. These systems learn about user interests over time and automatically suggest items that fit the learned model of user interests. It is important for companies to develop web-based marketing strategy such as product bundling to increase revenue. Recommendation system is a platform that can be used to reduce the searching cost of users, increase the effectiveness of promotion strategies and enhance loyalty. The core technology implemented behind this type of recommender systems includes content analysis, collaborative filtering and some hybrid variants. Collaborative filtering is a data analysis task appearing in many challenging applications and it can often be formulated as identifying patterns in a large and mostly empty rating matrix. In this paper, firstly, the principle of collaborative filtering recommendation is introduced. Then, describes the workflow of the collaborative filtering algorithm. Unresolved issues of collaborative filtering technology and research directions are pointed out finally.