Personalizing Group Recommendation to Social Network Users

Today, due to their flexibility and ease of use, social networks have fallen in the center of attention for users. The variety of social network groups has made users uncertain. This diversity has also made it difficult for them to find a group that well suits their preferences and personality. Therefore, to overcome this problem, we introduce the group recommendation system. This system offers customized recommendations based on each user's preferences. It is created by selecting related features based on supervised entropy as well as using association rules and D-Tree classification method. Assuming that members in each group share similar characteristics, heterogeneous members are identified and removed. Unlike other methods, this method is also applicable for users who have just been joined to the social network while they do not have friendship relationships with others or do not yet have memberships in any groups.