Improved recommendation algorithm based on bipartite networks
Duan Shuang-yan · Jisuanji yingyong yanjiu · 2013
In recent years,the recommendation algorithm based on networks has been attracting more and more researchers' attention.However,these recommendation algorithms based on bipartite networks are only to judge whether the user has selected the objects instead of distinguishing the preferences of the user about the object.And these algorithms tend to recommend popular objects,without considering the influence of object degree and the weights of the object.To solve these problems,this paper proposed an improved recommendation algorithm based on weighted networks,which distinguished the level of rating that a user voted an object.At the same time,the ratio θ of the object degree and the sum of weights of the object were embedded into the similarity index between users to improve the recommendation diversity.Experimental results show that the improved algorithm can improve recommendation accuracy and diversity,while reducing the epidemic of the recommended objects.