An efficient SVM-Based method to detect malicious attacks for web servers

Authors:
Wu Yang;Xiao-Chun Yun;Jian-Hua Li
Affiliations:
Information Security Research Center, Harbin Engineering University, Harbin, China;Information Security Research Center, Harbin Engineering University, Harbin, China;Information Security Research Center, Harbin Engineering University, Harbin, China
Venue:
APWeb'06 Proceedings of the 2006 international conference on Advanced Web and Network Technologies, and Applications
Year:
2006

Citing 8
Cited 0

State Transition Analysis: A Rule-Based Intrusion Detection Approach

IEEE Transactions on Software Engineering
The nature of statistical learning theory

The nature of statistical learning theory
Fast training of support vector machines using sequential minimal optimization

Advances in kernel methods
NATE: Network Analysis of Anomalous Traffic Events, a low-cost approach

Proceedings of the 2001 workshop on New security paradigms
Vector quantization based on genetic simulated annealing

Signal Processing
ADMIT: anomaly-based data mining for intrusions

Proceedings of the eighth ACM SIGKDD international conference on Knowledge discovery and data mining
Identifying key features for intrusion detection using neural networks

ICCC '02 Proceedings of the 15th international conference on Computer communication
A Neural Network Component for an Intrusion Detection System

SP '92 Proceedings of the 1992 IEEE Symposium on Security and Privacy

Quantified Score

Hi-index	0.00

Visualization

Abstract

In recent years, with the rapid development of network technique and network bandwidth, the network attacking events for web servers such as DOS/PROBE are becoming more and more frequent. In order to detect these types of intrusions in the new network environment more efficiently, this paper applies new machine learning methods to intrusion detection and proposes an efficient algorithm based on vector quantization and support vector machine for intrusion detection (VQ-SVM). The algorithm firstly reduces the network auditing dataset by using VQ techniques, produces a codebook as the training example set, and then adopts fast training algorithm for SVM to build intrusion detection model on the codebook. The experiment results indicate that the combined algorithm of VQ-SVM can greatly improve the learning and detecting efficiency of the traditional SVM-based intrusion detection model.