Recommender Systems have become an attractive field within the recent decade because they facilitate users' selection process within limited time. Conventional recommender systems have proposed numerous methods focusing on recommendations to individual users. Recently, due to a significant increase in the number of users, studies in this field have shifted to properly identify groups of people with similar preferences and provide a list of recommendations for each group. Offering a recommendations list to each individual requires significant computational cost and it is therefore often not efficient. So far, most of the studies impose four restrictive assumptions: (1) limited number of users, (2) number of groups, (3) average number of group members, and (4) full knowledge of the network topological structure. To overcome these limitations, we propose a novel approach which improves the accuracy of recommendations list to each group using network centrality concept. In this approach, the most central users are identified as heads of the groups, and then groups of users with similar preferences are consequently formed. After the group formation, a new group profiling strategy is provided to aggregate preferences of group members relative to their central-ities. Our approach is evaluated in different types of group recommender systems compared to several common strategies over the MovieLens-1M dataset. Experimental results demonstrate that our group formation and group profiling, based on the proposed user centrality measure, lead to more accurate recommendations list for each group.
