Trusted Rating For Sna Using Svd Model

Abstract

The main aim of the paper is to enhance the performance of a novel recommendation site for hotel by data mining.When we are about to view the rating of a hotel given by other people, we can’t rely accurately on it as there are many fake reviews. So we can overcome the problem by viewing only the ratings given by our friends and friends of friends in our trusted network. TrustSVD therefore created on top of a state of the art recommendation algorithm. By considering both the implicit and explicit influence of ratings the predictions are made. The proposed technique is used to merge with social trust information thus from that we can get the trusted network. Thus by using the trusted network reviews we can avoid the fake reviews. In case of cold start and data sparsity problem the friends of friends list is considered.