A Survey on Social Circle Influenced Personalized Recommendation System

In contemporary world, social networking sites are becoming an inseparable part of everyones life. With the advent and affection of online social networks, the social network based approach to recommendation has emerged. Recommender systems are the tools of choice to choose online information relevant to a given user. Users are interested to share their experiences through ratings, reviews, polls, blogs etc. which assists to recommend the items of user interest. Rapid growth of information generated by online social networks leads to increase in demand of efficient and effective recommender systems to give accurate results. Traditional recommendation techniques are limited because they do not consider factors of social relation in the social network for giving recommendation. The new intrinsic parameters of social network like personal interest, interpersonal interest similarities and interpersonal influence bring opportunities and challenges for recommender system to solve the cold start and sparsity problem of datasets more efficiently. This survey paper is to study various traditional recommendation techniques and main three social aspects, and how these factors are to be fused into a personalized recommendation model to give efficient recommendations to the user.