An IDS Alert Aggregation Method Based on Clustering

Article Preview

Abstract:

How to aggregate and reduce duplicated alerts is one of the most important tasks in IDSs. This paper proposed an alert aggregation method, which clustering similar alerts into a hyper alert based on category and feature similarity. For each feature we define an appropriate similarity function. The overall similarity is weighted by a specifiable expectation of similarity. Experiments on DARPA2000 data set have demonstrated the effectiveness of this method.

You might also be interested in these eBooks

Info:

Periodical:

Advanced Materials Research (Volumes 219-220)

Pages:

156-159

Citation:

Online since:

March 2011

Export:

Price:

Permissions CCC:

Permissions PLS:

Сopyright:

© 2011 Trans Tech Publications Ltd. All Rights Reserved

Share:

Citation:

[1] F. Aurel, F. Cuppens. Using an intrusion detection alert similarity operator to aggregate and fuse alerts. The 4th Conference on Security and Network Architectures, Bat sur Mer, France. Jun, (2005).

Google Scholar

[2] P. Ning, Y. Cui, and D. S. Reeves. Constructing attack scenarios through correlation of intrusion alerts (full version). TR-2002-13, North Carolina State University, (2002).

DOI: 10.1145/586110.586144

Google Scholar

[3] A. Valdes and K. Skinner. Probabilistic Alert Correlation[C]. In Proceedings of RAID 2001, Lecture Notes in Computer Science 2212, Berlin, Springer-Verlag, 54-68.

DOI: 10.1007/3-540-45474-8_4

Google Scholar

[4] A. Valdes and K. Skinner. Blue Sensors, Sensor Correlation, and Alert Fusion, in RAID 2000, Toulouse, France, October (2000).

Google Scholar

[5] MIT Lincoln Lab. 2000 DARPA intrusion detection scenario specific datasets. http://www.ll.mit.edu / IST/ideval/data/2000/2000_data_index.html, (2000).

Google Scholar