The importance of k-shell in discovering key nodes in complex networks

Abstract

Outlier detection is drawing more attention in recent years. It has a wide variety of applications, including network intrusion detection and event detection. A great deal of research has been done in this area, using spectrum or MDL (Minimum Description Length) as important tools to find some outliers. In this paper, we bring the k-shell into the outlier detection in complex networks, using the structural entropy as a way to measure the feature of the whole complex network. Through the experiment both on a synthetic network and a real world network, we give the importance of k-shell in discovering outliers in complex networks.