The Recommendation System Of A Distrust Propagation Algorithm Based On The Breadth First Traversal

With the continuous development of network technology,we are now in the information ocean world from a very scarce information age.In the face of this huge information,it is very difficult to get quick access to the information they want.The recommendation system arises at the historic moment,can provide people with personalized network services,can quickly and accurately locate the information they need when people browse the web page.Recommendation system was first applied in the field of e-commerce,when the user to buy goods can get a good recommendation to enhance the customer experience.However,due to the problem of data sparsity and cold start,some users can not be well recommended.A recommendation system based on social network was born.In this algorithm,the trust relationship in social network is considered in the recommendation system.For example,user item rating matrix can be selected with trust relation of trust users,calculating similarity and selecting neighbor.Some of them is combining the similarity with the trust value together to form a new value to select the nearest neighbor etc..Either way,the efficiency and accuracy of the recommendation system has been improved,and the effect of the long tail effect has become smaller.But we also know that the social network is a complex network of social relations,which not only has a trust relationship,there are many factors such as trust relationship,so distrust in recommendation algorithm will factor into account is also granted.Although the current stage has some scholars to research this problem,but they are not comprehensive enough and did not give a complete trust model,finally they no data did not solve the sparsity problem of distrust relationship in the network.In order to solve the above problems,it is possible to have a better personalized recommendation algorithm to serve the users.The main work of this paper is as follows:We propose a new.In this paper,we propose a distrust propagation algorithm,which can be used to expand the distrust relationship of social networks.To approve distrust propagation algorithm based on the breadth first traversalproposed above,this paper designs distrust propagation model,the distrust relationship can be quantified.The distrust propagation algorithm based on the breadth first traversal combined with collaborative filtering algorithm based on trust,resulting in The Recommendation system of a distrust propagation algorithm based on the breadth first traversal.This algorithm can solve the problem of the sparsity of the distrust matrix in the social network,and also can alleviate the sparsity of the recommendation system.As Epinions data set for experiment,the collaborative filtering algorithm is analyzed and compared with the traditional collaborative filtering recommendation algorithm,confirmed that the new algorithm is superior to the traditional recommendation in effect.