Robust Data Clustering Algorithms for Network Intrusion Detection
Journal Title: International Journal of Computer & organization Trends(IJCOT) - Year 2012, Vol 2, Issue 5
Abstract
IDS (Intrusion Detection system) is an active and driving defense technology. Intrusion detection is to detect attacks against a computer system. This project mainly focuses on intrusion detection based on data mining. Data mining is to identify valid, novel, potentially useful, and ultimately understandable patterns in massive data. One of the primary challenges to intrusion detection are the problem of misjudgment, misdetection and lack of real time response to the attack. In the recent years, as the second line of defense after firewall This project presents an approach to detect intrusion based on data mining frame work. In this framework, intrusion detection is achieved using clustering techniques. Firstly, a method to reduce the noise in the data set using improved kmeans. This system use K-means,FCM and Improved K-means data mining algorithms are used to improves the performance of intrusion detection since the traffic is large and the types of attack are various. By the more accurate method of finding k clustering center, an anomaly detection model was presented to get better detection effect. This project used KDD CUP 1999 data set to test the performance of the model. The results show thesystem has a higher detection rate and a lower false alarm rate, it achieves expectant aim.
Authors and Affiliations
Gunja Ambica #1 , Mrs. N. Rajeswari#2
Keywords
Related Articles
- EP ID EP140961
- DOI -
- Views 91
- Downloads 0
A K-Anonymity Privacy-Preserving Location Monitoring System for Wireless Sensor Networks with Nymble Secure System
Anonymizing wireless sensor networks allow users to access services privately by using a series of routers to hide the client’s IP address from the server. As a result, administrators block all known exit nodes of...
IP Anycast Architecture
This paper illustrates the methodology and architecture for network addressing and routing in which datagram packets routed through mathematical topological nearest node in a cluster of potential receivers that a...
A Brief Survey On Document Clustering Techniques Using MATLAB
Document clustering is a more specific technique for unsupervised document organization, it is generally considered to be a centralized process. Clustering methods can be used to automatically group the retrieved documen...
Cataloguing and Avoiding the Buffer Overflow Attacks in Network Operating Systems
The application software has a different dimension, size and intricacies is rising rapidly in current technology era and simultaneously increase a programming bugs also. The programming bugs cause vulnerabilities to the...
Gradation of NGO’s Role in Rural Development: A Fuzzy Soft Set Theoretic Approach
The non-governmental organizations (NGO) are recognized as important institutional actors that are expected to facilitate the participation of beneficiaries in the development programmes such as mobilising communit...