Item Correlation Graph Based Collaborative Filtering Algorithm
Xiaoru Wang · 2014
In e-commerce,accurate recommendation can improve the trading volume,and thus bring more profit for enterprises.In order to improve the accuracy of recommender system,this paper proposed an item correlation graph based collaborative filtering algorithm.This paper assumed the items as nodes,the number of people buying two commodities as the weight of the edge,and constructed an item correlation graph.According to the item correlation graph,this paper proposed a recommender algorithm that takes both item correlation graph based similarity and average similarity into consideration.Experiments show that the proposed algorithm has better prediction accuracy,and is more effective than related algorithms.