To search, Click below search items. |
| All Published Papers Search Service |
|
|
|
Title |
Intrusion Detection System Based on Conditional Random Fields |
|
Author |
Deepa V. Guleria, Chavan M.K |
| Citation |
Vol. 13 No. 12 pp. 80-86 |
|
Abstract |
An intrusion detection system is used to monitor network traffic, check for suspicious activities and notifies the network administrator or the systemTo operate in high speed networks, present network intrusion detection systems are either signature based or anomaly based system. These systems are inefficient and suffer from a large number of false alarms. Some of the common attacks such as DoS,R2L ,Probe and U2R affect the network resources. Intrusion detection system has challenges to detect malicious activities reliably and should able to perform efficiently with large amount of network traffic.We address in this paper two major issues of Accuracy and Efficiency by introducing a probabilistic approach Conditional Random Fields and Sequential Layered Approach.It is demonstrated that using Conditional Random Fields high attack detection accuracy can be achieved and using the Sequential Layered Approach high efficiency. Our experimental results on the benchmark KDD 1999 intrusion data set show improvement in attack detection accuracy is very high for Probe, Denial of Service, U2R and R2L attacks. |
|
Keywords |
Intrusion Detection, Conditional Random Fields, Network Security, Decision tree |
|
URL |
