Multi stage attack Detection system for Network Administrators using Data Mining

Conference · Thu Apr 01 00:00:00 EDT 2010

OSTI ID:986832

Rajeshwar, Katipally ^[1]; Gasior, Wade C ^[2]; Yang, Dr Li ^[1]

---

1. University of Tennessee
2. ORNL

In this paper, we present a method to discover, visualize, and predict behavior pattern of attackers in a network based system. We proposed a system that is able to discover temporal pattern of intrusion which reveal behaviors of attackers using alerts generated by Intrusion Detection System (IDS). We use data mining techniques to find the patterns of generated alerts by generating Association rules. Our system is able to stream real-time Snort alerts and predict intrusions based on our learned rules. Therefore, we are able to automatically discover patterns in multistage attack, visualize patterns, and predict intrusions.

• Research Organization: Oak Ridge National Lab. (ORNL), Oak Ridge, TN (United States)
• Sponsoring Organization: Work for Others (WFO)
• DOE Contract Number: DE-AC05-00OR22725
• OSTI ID: 986832
• Resource Relation: Conference: Cyber Security and Information Intelligence Research Workshop, Oak Ridge, TN, USA, 20100421, 20100423
• Country of Publication: United States
• Language: English

Related Subjects

99 GENERAL AND MISCELLANEOUS//MATHEMATICS, COMPUTING, AND INFORMATION SCIENCE
DETECTION
INTRUSION DETECTION SYSTEMS
SECURITY
DATA ACQUISITION SYSTEMS
COMPUTER NETWORKS