A Weighted Slope One Algorithm Based on Cluster Filling and Time Weight

Abstract

In order to solve defects of the Slope One algorithm that the effect of recommending is not well because of without considering the time weight, and has the problem of data sparsity and poor real-time performance. A weighted slope one algorithm based on cluster filling and time weight (WSOBCFT) was proposed in this paper. To reduce the time of generating the nearest neighbor, the rating matrix of time weight was clustered by Canopy-K-means algorithm at first. Then every class was filled to improve the density of the matrix and reduce data sparsity. Finally, considering the similarity weight and time weight, item's rating was predicted in the set of nearest neighbor. The experimental results illustrate that compared with traditional recommendation algorithm, the proposed algorithm has higher accuracy and real-time performance.