A survey on group modeling strategies for recommender systems

Abstract

Abstract E-commerce websites generate huge volumes of customer review data. In this era of bigdata, recommender systems have played a vital role in suggesting customized recommendations to the users. Recommender systems generally deal with user preferences. Increase in online group activities has fueled the need for group recommender system extremely. A group recommender system recommends one item (or more) to a group of users considered together (with some matching similarity). Group recommender systems are used in various domains like music, movies, travel, etc. Group recommendation uses the preference of their group members in order to recommend the items to a group of users. This chapter explores various group modeling strategies, their implementation methods, and modified variations. Some of the group modeling techniques are additive, least misery, Borda count, Copeland, most pleasure, most respected, mean after threshold, fairness, etc. All the modeling techniques are explained with suitable examples and solved problems.