Comparison of Group Recommendation Techniques in Social Networks

Virtual communities and groups are known as one of the features of social networks for creating the possibility for users to join together and interact. Regarding the growth of social networks as well as attracting new users of various ages and creation of different groups, assisting users seems quite necessary. Along with studying some of the common recommender methods in social networks in this paper, a new method is explained. This new method is designed using d-tree classification, association rules and the concepts of information theory which compared with others, it gives better results. It is also possible in this system to offer recommendations to new users who have just joined the network and do not have any links.