期刊名称：International Journal of Computer Information Systems and Industrial Management Applications
印刷版ISSN：2150-7988
电子版ISSN：2150-7988
出版年度：2014
期号：6
页码：515-527
出版社：Machine Intelligence Research Labs (MIR Labs)
摘要：Providing Group Recommendations is an openresearch area. In the proposed scheme, the combined entropybased similarities using positive and negative preference ratingsamong training users are used to extract Similar Taste Users(STUs). Such STUs build a group for the target user from whichgroup recommendations are generated. Using information gain,it further computes top N individual recommendations fromthese group recommendations based on opposite userpreferences.In this paper, a method is proposed to overcome the sparsityamong preferences of group members. Genre based similarity(based on implicit multi criteria information) among target userand each group member generates genre based profile of targetuser which in turn increases the density of preferences amonggroup members. Movie Lens dataset is used for experiments. Itshows significant improvements in overcoming sparsity problemin group recommender systems and performance measures usedshows improvement in recommendation quality.
关键词：Group Recommendation; Sparsity Problem; Genre;Based Similarity; Entropy; Information Gain; User Rating;Preferences.
