Variance-Index Based Feature Selection Algorithm for Network Intrusion Detection
Journal Title: IOSR Journals (IOSR Journal of Computer Engineering) - Year 2016, Vol 18, Issue 4
Abstract
Abstract: This paper presents a feature selection methodology in the domain of Network Intrusion Detection System (NIDS). An Unsupervised Variance Indexed Feature Selection Algorithm (VIFS) is proposed and demonstrated by considering a benchmark dataset, NSL KDD. A Fast KNN classification algorithm Indexed PartialDistance Search k Nearest Neighbor(IKPDS) is applied for finding the classification accuracy. The algorithm selects a subset of features based on fitness threshold value with tolerable loss in classification accuracy but gain in computational time. In a trade of NIDS classification accuracy and computational time are two factors to decide the performance of NIDS. Feature subset length is one of the parameter to influence the computational time. So in order to evaluate the fitness value to consider the classification accuracy and feature subset length in this study. Two parameters α and β are related to the presence of classification accuracy quality and feature subset length. A numerical illustration presented for 0.5 ≤ α ≤ 9, where α =[0,1] and β =1- α. Then identified three feature selection scenarios. Finally the merits and demerits of these three scenarios are discussed. This VIFS fulfills the gain in computational time objective with a tolerable loss in classification accuracy.
Authors and Affiliations
B. Basaveswara Rao , K. Swathi
Keywords
Related Articles
- EP ID EP90837
- DOI -
- Views 104
- Downloads 0
Secure Dispatch of Mobile Sensors in a Hybrid Wireless Sensor Networks
A Hybrid Wireless Sensor network (HWSN) with static and mobile nodes is considered. Static sensors monitor the environment and report events occurring in the sensing field. Mobile sensors are then dispatched to v...
Palm Vein Extraction and Matching For Personal Identification
Abstract : The most common method used for authentication is Textual passwords. But Textual passwords are in risk to eves dropping, dictionary attacks, social engineering and shoulder surfing. Biometric passwords a...
Optimizing Neuro-Fuzzy Fault Diagnostic Algorithm for Photovoltaic Systems
Abstract: The main goal of this research is to develop a novel optimum neuro-fuzzy system for diagnosis the complex and dynamic systems. .It has used the Particle Swarm Optimization (PSO) technique for training the Adapt...
Sentiment Mining and Related Classifiers: A Review
Abstract: Brokers trading goods on the Web often seek for their customer’s reviews and feedbacks of their products and the associated services. With the growing popularity of e-commerce, the number of customer opinions t...
Software Development Process by a Logical Approach to Quantify the Throughput by Balancing Time and Cost
Abstract: Q-SPIM is an organized advance and incessant progress to enhance software organization’s ability throughput quality software that balances time and cost [2]. Such a model developed is verified and validated by...