Algorithm for tuning fuzzy network attack classifiers based on invasive weed optimization

Author

Anfilofiev, A.E. ; Hodashinsky, I.A. ; Evsutin, O.O.

Author_Institution

Complex Inf. Security of Comput. Syst. Dept., Tomsk State Univ. of Control, Tomsk, Russia

fYear

2014

fDate

11-13 Nov. 2014

Firstpage

1

Lastpage

4

Abstract

The purpose of this work is to describe a hybrid approach for constructing intrusion detection systems that incorporates feature extraction algorithms and algorithms for tuning classifiers. In this paper, we construct the classification algorithm based on the invasive weed optimization algorithm and use the genetic algorithm (GA) to reduce the dimension of the feature space. The experimental results support the efficiency of the proposed approach for solving intrusion detection problems.

Keywords

feature extraction; genetic algorithms; pattern classification; security of data; GA; feature extraction algorithms; fuzzy network attack classifier tuning algorithm; genetic algorithm; intrusion detection systems; invasive weed optimization algorithm; Accuracy; Algorithm design and analysis; Classification algorithms; Genetic algorithms; Intrusion detection; Support vector machine classification; Training; DARPA dataset; feature selection; fuzzy systems; genetic algorithm; intrusion detection; learning classification rules; network security; pattern recognition; swarm intelligence;

fLanguage

English

Publisher

ieee

Conference_Titel

Dynamics of Systems, Mechanisms and Machines (Dynamics), 2014

Conference_Location

Omsk

Print_ISBN

978-1-4799-6406-2

Type

conf

DOI

10.1109/Dynamics.2014.7005632

Filename

7005632