Analyzing the effectiveness of graph metrics for anomaly detection in online social networks

Authors:
Reza Hassanzadeh;Richi Nayak;Douglas Stebila
Affiliations:
School of Electrical Engineering and Computer Science, Science and Engineering Faculty, Queensland University of Technology, Brisbane, Australia;School of Electrical Engineering and Computer Science, Science and Engineering Faculty, Queensland University of Technology, Brisbane, Australia;School of Electrical Engineering and Computer Science, Science and Engineering Faculty, Queensland University of Technology, Brisbane, Australia
Venue:
WISE'12 Proceedings of the 13th international conference on Web Information Systems Engineering
Year:
2012

Citing 4
Cited 1

Neighborhood Formation and Anomaly Detection in Bipartite Graphs

ICDM '05 Proceedings of the Fifth IEEE International Conference on Data Mining
Measurement and analysis of online social networks

Proceedings of the 7th ACM SIGCOMM conference on Internet measurement
Power-Law Distributions in Empirical Data

SIAM Review
OddBall: spotting anomalies in weighted graphs

PAKDD'10 Proceedings of the 14th Pacific-Asia conference on Advances in Knowledge Discovery and Data Mining - Volume Part II

A semi-supervised graph-based algorithm for detecting outliers in online-social-networks

Proceedings of the 28th Annual ACM Symposium on Applied Computing

Quantified Score

Hi-index	0.00

Visualization

Abstract

Online social networks can be modelled as graphs; in this paper, we analyze the use of graph metrics for identifying users with anomalous relationships to other users. A framework is proposed for analyzing the effectiveness of various graph theoretic properties such as the number of neighbouring nodes and edges, betweenness centrality, and community cohesiveness in detecting anomalous users. Experimental results on real-world data collected from online social networks show that the majority of users typically have friends who are friends themselves, whereas anomalous users' graphs typically do not follow this common rule. Empirical analysis also shows that the relationship between average betweenness centrality and edges identifies anomalies more accurately than other approaches.