A new grouping method based on social choice strategies for group recommender system

Abstract

Recommender System is a software or tool that helps users to select items or things according to their preferences. These are used in almost every web sites today. A lot of research is going on, how to produce efficient recommendations for individuals. Even more, today the recommendation of items/things are for a group of users where there is more than one user in a group and each user have their own preferences. Group Recommender System recommends items or things for a group of users based on their individual preferences. There are many social choice grouping strategies available. We proposed a new grouping algorithm which will first generate homogeneous groups and then generate recommendation of items for them. In this paper we followed the rule based approach to learn the user’s preferences. All the results of our approach is validated with the movie lens data set which is the bench mark data set for recommender system testing.