A dynamic network anomaly detection method based on trend analysis

Detecting abnormal behavior during network evolution is an important and challenging data analysis task at present, it is named as dynamic network anomaly detection in this paper. The network abnormal behavior differs from that of normal behavior, the probability of occurrence stays relatively low but may cause serious damages once happened. This paper takes the topological structure of the network as the research object and identifies the network anomaly through the change of network structure. Based on Cox-stuart test, we put forward a new type of method to detect dynamic anomaly. This method refers to an alarm of monitoring the trend change of the network structure parameters. Finally, an experiment was carried out in the real dynamic network-Enron Email Network, which verified that the dynamic network anomaly detection method proposed in this paper can effectively detect the network anomaly behavior.

[1]  David J. Marchette,et al.  Scan Statistics on Enron Graphs , 2005, Comput. Math. Organ. Theory.

[2]  Alan Veliz-Cuba,et al.  Steady state analysis of Boolean molecular network models via model reduction and computational algebra , 2014, BMC Bioinformatics.

[3]  Jie Li,et al.  Social network change detection using a genetic algorithm based back propagation neural network model , 2016, 2016 IEEE/ACM International Conference on Advances in Social Networks Analysis and Mining (ASONAM).

[4]  Kathleen M. Carley,et al.  Measuring Temporal Patterns in Dynamic Social Networks , 2015, ACM Trans. Knowl. Discov. Data.

[5]  Amany Sarhan,et al.  A hybrid outlier detection algorithm based on partitioning clustering and density measures , 2015, 2015 Tenth International Conference on Computer Engineering & Systems (ICCES).

[6]  Christopher Leckie,et al.  High-dimensional and large-scale anomaly detection using a linear one-class SVM with deep learning , 2016, Pattern Recognit..

[7]  Yunfei Li,et al.  Research on outlier detection for high dimensional data stream , 2016 .

[8]  Danai Koutra,et al.  Graph based anomaly detection and description: a survey , 2014, Data Mining and Knowledge Discovery.

[9]  Farshad Safaei,et al.  A statistical infinite feature cascade-based approach to anomaly detection for dynamic social networks , 2017, Comput. Commun..

[10]  Nirvana Meratnia,et al.  A taxonomy framework for unsupervised outlier detection techniques for multi-type data sets , 2007 .

[11]  Pengfei Chen,et al.  An Automatic Framework for Detecting and Characterizing Performance Degradation of Software Systems , 2014, IEEE Transactions on Reliability.

[12]  Aric Hagberg,et al.  Exploring Network Structure, Dynamics, and Function using NetworkX , 2008 .

[13]  Kathleen M. Carley.  ORA: A Toolkit for Dynamic Network Analysis and Visualization , 2014, Encyclopedia of Social Network Analysis and Mining.

[14]  Kathleen M. Carley.  Dynamic Network Analysis , 2003 .

[15]  Yizhou Sun,et al.  Integrating community matching and outlier detection for mining evolutionary community outliers , 2012, KDD.