Type-based context-aware service Recommender System for social network
Back Sun Sim, Heeseong Kim, Kwang Myung Kim, Hee Yong Youn · 2012
As social network grows fast recently, numerous services are provided based on it. Here it is difficult for the users to choose a right service if a large number of choices exist. Recommender System (RS) is a software tool providing the suggestions of the items most useful to the user. The existing RSs do not properly consider the social relation of the entities in rating the choices. This paper proposes a new context-aware recommendation scheme which reflects the user type in estimating the closeness between the users in social network, together with the cosine similarity measure. Computer simulation shows that the proposed scheme significantly improves the accuracy of rating compared to the existing ones employing the multi-dimensional paradigm and rough set theory. It also displays that the amount of items recommended by the proposed approach is larger than with the existing approaches.