Title Of Paper:
 
 
Intrusion Detection in Computer Networks using a Fuzzy-Heuristic Data Mining Technique
Author's Name :  Hamid Saadi, MohamadBagherKeley
KeyWords:  Simulated Annealing, Fuzzy Systems, Fuzzy Rule Extraction, Intrusion Detection, Pattern Classification
Pages:  25 -38
Volume: 3
Issue: 12
Year: 2015
Abstract:
 

In this article the use of Simulated Annealing (SA) algorithm for creating a consistent intrusion detection system is presented. The ability of fuzzy systems to solve different types of problems has been demonstrated in several previous studies. Simulated Annealing based Fuzzy Intrusion Detection System (SAF-IDS) crosses the estimated cognitive method of fuzzy systems with the learning capability of SA. The objective of this paper is to prove the ability of SAF-IDS to deal with intrusion detection classification problem as a new real-world application area which is not previously undertook with SA-based fuzzy system. Here, the use of SA is an effort to efficiently explore and exploit the large examines space usually related with the intrusion detection problem, and finds the optimum set of fuzzy if-then rules.  The proposed SAF-IDS would be capable of extracting precise fuzzy classification rules from network traffic data and relates them to detect normal and invasive actions in computer networks. Tests were performed with KDD-Cup99 intrusion detection benchmark which is widely used to calculate intrusion detection algorithms. Results indicate that SAF-IDS provides more accurate intrusion detection system than several well-known and new classification algorithms.

Full Text: