PRBL

In recent years complex network has become a research hot spot for the large-scale systems. The bipartite network, a particular format of the complex network, can be used to describe the personalized recommendation systems that recommend the interesting items to users based on their own interests. In this paper, we propose a personalized recommendation systems based on the bipartite network projection and link community detection. The preference from users to items is abstracted as the edges in the bipartite user-item network. To get the relationship between the items, we make the one-mode projection to generate the item-item network. The link community detection is used to cluster the items to recommend to preferred users. Our systems could be used for any area with large-scale datasets. The experimental results show that our system could efficiently recommend commodities to the consumers based on the big data from the e-commerce.