Research On Personalized Recommendation Systemfor Group Purcuase Websites

With the development of the Internet, E-business has developed rapidly since21century. In recent years, a new E-business pattern called’group purchase’develops rapidly, attracts many consumers in a short time, and has become one kind of the most important E-business patterns. As the competition in group purchase websites become fiercer, how to provide better service has become an important subject. The problem is, there are many goods on sale in group purchase websites every day. Users have to spend long time browsing all the item pages to find what they are interested in. We consider using personalized recommendation system which was proposed at the end of last century and has been applied in many domains to solve the problem above.This paper introduces several popular personalized recommendation methods, makes comparisons in advantages, disadvantages and application areas between them. We choose collaborative filtering method as the core recommendation method in our system after comparison. A new method to compute similarity between items has been proposed in this paper to improve the effect of our recommendation, which has proven to be effective.After studying the features of group purchase websites, we design a personalized recommendation system for group purchase websites to improve the effect of recommendation. This paper proposes a recommendation method which combines users’this day’s log and former log, and discusses the ways to process log, generate recommendation strategy, generate recommendation list.