Abstract
Whenever an intrusion happens, privacy and security of the system are compromised. In order to detect different types of attacks that happen in a network, Intrusion Detection System (IDS) plays a crucial role in Network security. IDS is designed in order to classify the activities of the system into abnormal and normal. Machine learning based Intrusion Detection is gaining attention in recent years and is able to give better results with greater accuracy and high detection rate on novel attacks. In this paper, performance of different kernels of Support Vector Machine (SVM) are evaluated against Knowledge Discovery in Databases Cup'99(KDD) data set and detection accuracy, detection time are compared. The detection time is reduced by adopting Principal Component Analysis (PCA) which curtails higher dimensional dataset to lower dimensional dataset. The experiments which are conducted in this research shows that Gaussian Radial Basis Function kernel of SVM has higher detection accuracy.
Original language | English |
---|---|
Title of host publication | 2016 IEEE International Conference on Recent Trends in Electronics, Information and Communication Technology, RTEICT 2016 - Proceedings |
Publisher | Institute of Electrical and Electronics Engineers Inc. |
Pages | 1344-1350 |
Number of pages | 7 |
ISBN (Electronic) | 9781509007745 |
DOIs | |
Publication status | Published - 05-01-2017 |
Externally published | Yes |
Event | 1st IEEE International Conference on Recent Trends in Electronics, Information and Communication Technology, RTEICT 2016 - Bangalore, India Duration: 20-05-2016 → 21-05-2016 |
Conference
Conference | 1st IEEE International Conference on Recent Trends in Electronics, Information and Communication Technology, RTEICT 2016 |
---|---|
Country/Territory | India |
City | Bangalore |
Period | 20-05-16 → 21-05-16 |
All Science Journal Classification (ASJC) codes
- Computer Networks and Communications
- Computer Science Applications
- Information Systems
- Electrical and Electronic Engineering