Modeling the preferences of a group of users detected by clustering: A group recommendation case-study

Group modeling is the process that combines multiple user models into a single model. In group recommendation, this allows to derive a group preference for each item. It is known that the strategy used to model a group has to be chosen considering the domain in which the system operates. This paper evaluates group modeling strategies in a group recommendation scenario in which groups are detected by clustering users. Once users are clustered, strategies are tested, in order to find the one that allows to get the best accuracy. Experimental results show that clustering and group modeling are strongly connected. By producing group preferences that are equally distant from the individual preferences, the modeling strategy has the same role that the centroid has when users are clustered. This previously unknown link among the two tasks is essential in order to build accurate group recommendations.