Recommender Network using Indirect Social Relation Matrix Factorization and Social Latent Factors

Abstract

Recommendation system provide suggestion for the user based on their personal preference or influence from another user. There are four real - world datasets (Ciao, Epinions, Douban, Yelp) that will be used to perform the demonstration capabilities of the proposed model . This results will improve the performance of the recommender system. Indirect social relationships are more capable due to systematic quality recommendations. In this paper, we present a new general model of indirect social relationship detection and matrix factorization collaborative filtering technology in community system. Possible elements can simultaneously capture the personal choice , favouritism and behavior of a user's community group.