| Publisher | University of Regina | ||
|---|---|---|---|
| Format | 173.5KB PDF | Date added | 09 Aug 2006 |
| Topics | Network Security, Security Tools, Intrusion Detection Systems | ||
| Downloads | 32 | ||
Design and implementation of intrusion detection systems remain an important research issue in order to maintain proper net-work security. Support Vector Machines (SVM) as a classical pattern recognition tool has been widely used for intrusion detection. However, conventional SVM methods do not concern different characteristics of features in building an intrusion detection system. This paper proposes an enhanced SVM model with a weighted kernel function based on features of the training data for intrusion detection. Rough set theory is adopted to perform a feature ranking and selection task of the new model. The new model is evaluated with the KDD dataset and the UNM dataset. It is suggested that the proposed model outperformed the conventional SVM in precision, computation time, and false negative rate.
Did you find this white paper useful?
1 out of 4 users found this white paper useful
Related white papers
MessageLabs Intelligence : 2009 security Predictions
Having analyzed the global threat landscape for almost a decade, MessageLabs Team Skeptic™ is comprised of many world-renowned malware and spam experts who have a global view of threats across...
IDC Vendor Spotlight
Organised ubiquity is a must for organisations to sucessfully "project" their users in any given landspace, at any given time, with secuirty policy. This White Paper covers issues surrounding secure...
Trend Micro Enterprise Security white paper
This white paper reviews the content security threat landscape and how it has evolved into a more dangerous and high risk environment. The paper discussed how conventional content security approaches...
Smart Protection E-Book
"Outthink the Threat: Why conventional protection is no match for new data-stealing malware & the Trend Micro Smart Protection Network is." Discover how cyber criminals are rendering traditional security solutions...
Secure Desktop On-Demand Webcast
The desktop or endpoint is one of the most vulnerable parts of your environment. Threats are everywhere. You have users who love to experiment with device settings (only to wonder...
Trend Micro Web Threats WP
Web threats pose a broad range of risks, including financial damages, identity theft, loss of confidential business information, theft of network resources, damaged brand or personal reputation, and erosion of...
The Past, Present and Future of Whitelisting
The explosion in malware and zero-day vulnerabilities over the last several years has limited the effectiveness of blacklisting technology. While still important, it must be augmented with other technologies such...
