Font Size: a A A

Research Of Recommender System Based On Limited Attention

Posted on:2019-09-24Degree:MasterType:Thesis
Country:ChinaCandidate:Y Z LiuFull Text:PDF
GTID:2428330548979756Subject:Computer technology
Abstract/Summary:PDF Full Text Request
With the Internet rapidly developing,information overload occurs.A massive amount of data has been produced in our daily life,but user's time and energy is limited.It becomes a very difficult thing to find the interested information from large volumes of data.Over the past decade,the recommender system has been extensively studied and applied in various fields.In the traditional recommender system,the user's social relations,interest classification and trust circle are added to improve the recommendation performance.But the current recommender system often neglects the psychological factors which play an important role in our behaviors.And one of the very important factor is the user's attention.In this paper,a general method is provided by incorporating limited attention information to improve recommender system.The main work of this paper consists of three aspects:(1)generally describe the idea of social recommender systems,including the algorithms based on collaborative filtering and social networks,and the matrix factorization model;(2)specifically describe how to improve the recommender system with limited attention information.By using social regularization term indicates the restriction that the limited attention acts on the recommender system,designs the objective function of the matrix factorization.Then,using the relevance of the item,propose the socialization of the item and make it as the regularization item to incorporate into the objective function.(3)set up a few comparative experiments and run them on different datasets to predict the score.The results show that the proposed method is superior to the mean absolute error MAE,root mean square error RMSE,and other indicators Social recommendation system.
Keywords/Search Tags:Recommender System, Matrix Factorization, Social Regularization
PDF Full Text Request
Related items