Fusion of Statistic, Data Mining and Genetic Algorithm for feature selection in Intrusion Detection
Journal Title: International Journal of Advanced Research in Computer Engineering & Technology(IJARCET) - Year 2013, Vol 2, Issue 5
Abstract
The security of information and data is a critical issue in a computer networked environment. In our society computer networks are used to store proprietary information and to provide services for organizations and society. So in order to secure this valuable information from unknown attacks (intrusions) need of intrusion detection system arises. There are many intrusion detection approaches focused on the issues of feature reduction as some of the features are irrelevant or redundant which results in lengthy detection process and degrading the performance of an IDS. So in order to design lightweight IDS we investigate the performance of three feature selection approaches CFS, Information Gain and Gain Ratio. In this paper we propose a fusion model by making use of the three standard algorithms and finally applying genetic algorithm that identify important reduced input features. We apply Naive Bayes classifier on the dataset for evaluating the performance of the proposed method over the standard ones. The reduced attributes shows that proposed algorithm give better performance that is efficient and effective for detecting intrusions
Authors and Affiliations
MeghaAggarwa , Amrita
Keywords
Related Articles
- EP ID EP115208
- DOI -
- Views 89
- Downloads 0
Design and Implementation of Structured and Unstructured Data Querying System in Heterogeneous Environment
In real world applications, data sharing and integration system design is challenging task with the problem. The system currently is only used to integrate and query on structured data (e.g., data of a relational databas...
Routing Issues for Trust Based Framework in Mobile Ad Hoc Networks
The execution and survival of ad hoc networks depends on cooperative and trusting nature of the distributed nodes. There is a common assumption in the routing protocols that all nodes are trustworthy and cooperative....
Optimization of Traffic System using TCL Algorithm through FMSA and IMAC Agents
Based on Agent Technology, Service Oriented Architecture and RFID,this paper puts forward a ‘real time paradigm’ traffic congestion control strategy in metropolitan cities. This strategy uses TCL algorithm in order...
Human Identification Based Biometric Gait Features using MSRC Human Identification Based Biometric Gait Features using MSRC
Human identification using gait is a new biometric intended to classify individuals by the way they walk have come to play an increasingly important role in visual surveillance applications. In gait biometric resea...
A General Method to Solve All Two-Port Ladder Networks
Solving the two-port ladder networks to determine the parameters such as the driving point impedance, the transfer impedance and the voltage transfer ratio requires writing the given network in the form of KVI an...