Towards Social Big Data-Based Affective Group Recommendation

Abstract

Social big data is currently an emergent issue, especially for recommender systems. In particular, with respect to social big data, various data mining techniques have been applied in group recommender systems. However, three social phenomena (i.e., social influence, emotional contagion, and conformity) have not been applied enough in existing studies. In this paper, a novel method for a group recommendation is proposed based upon the affective social network from social big data. In this regard, to explore and measure the social phenomena, a variety of similarity measures were applied in a content-based recommendation. Moreover, the proposed method has a computational complexity of O(N (2)), where N is the number of users in a group. Therefore, it is appropriate for big data environments, since the N is generally small for user groups. This study's results revealed that the Mahalanobis distance was suitable for the affective group recommendation. Moreover, the proposed method outperformed the other group recommender systems, those with large groups.