As a guest user you are not logged in or recognized by your IP address. You have
access to the Front Matter, Abstracts, Author Index, Subject Index and the full
text of Open Access publications.
Anomaly detection on graphs of social or communication networks has important security applications. The definition of a graph anomaly typically depends on the data and application of interest. Usually central in anomaly detection are the connections amongst the graph's entries and various methods have been developed for their analysis. We review these techniques and also discuss challenges currently faced for high-dimensional dynamic graphs.