An Outlying Reduction Algorithm Based on Power Graph

Abstract

This paper puts forward an outlying reduction method based on the power graph. The outlying reduction problem is due to the search of power graph. This paper utilizes the pruning strategy on the basis of the power graph expansion, which greatly reduces the storage and computation and improves the algorithm performance. This method can explain and analyze outliers in a smallest outlying subspace which contains the most outliers. The algorithm is illustrated by an example and the experimental results show the feasibility and efficiency of the algorithm.