Filter-based algorithms for implementing an intrusion detecting system during KDD
Author(s):
K.Vivek, G.Ananthnath
Keywords:
Intrusion detection, Least square support vector machine, Feature selection.
Abstract
Redundant and unsuitable features in data have caused a long-run drawback in network traffic classification. These features not solely slow down the method of classification however also stop a classifier from creating accurate decisions, particularly once coping with big data. In this paper, we tend to propose a mutual data based mostly algorithmic program that analytically selects the optimum feature for classification. This mutual data based feature selection algorithmic program will handle linearly and nonlinearly dependent data options. Its effectiveness is evaluated within the cases of network intrusion detection. an Intrusion Detection System named Least sq. Support Vector Machine based mostly IDS (LSSVM-IDS), is built using the features chosen by our proposed feature choice algorithmic program. The performance of LSSVM-IDS is evaluated using 3 intrusion detection analysis datasets, specifically KDD Cup 99, NSL-KDD and Kyoto 2006+ dataset. The analysis results show that our feature choice algorithm contributes a lot of critical options for LSSVM-IDS to attain higher accuracy and lower process value compared with the state-of the-art methods.
Article Details
Unique Paper ID: 145534
Publication Volume & Issue: Volume 4, Issue 10
Page(s): 415 - 419
Article Preview & Download
Share This Article
Join our RMS
Conference Alert
NCSEM 2024
National Conference on Sustainable Engineering and Management - 2024