An Intelligent Hybrid GA-PI Feature Selection Technique for Network Intrusion Detection Systems

Title

An Intelligent Hybrid GA-PI Feature Selection Technique for Network Intrusion Detection Systems

Creator

Sowmya; Mary Anita T.

Description

The development of Network Intrusion Detection Systems (NIDS) has become increasingly important due to the growing threat of cyber-attacks. However, with the vast amount of data generated in networks, handling big data in NIDS has become a major challenge. To address this challenge, this research paper proposes an intelligent hybrid GA-PI algorithm for feature selection and classification tasks in NIDS using support vector machines (SVM). The proposed approach is evaluated using two sub-datasets, Analysis and Normal, and Reconnaissance and Normal, which are generated from the publicly available UNSWNB-15 dataset. In this work, instead of considering all possible attacks, the focus is on two attacks, emphasizing the importance of the feature selection agent in determining the optimal features based on the attack type. The experimental results show that the proposed hybrid feature selection approach outperforms existing methodologies in terms of accuracy and execution time. Moreover, the selection of features can be subjective and dependent on the domain knowledge of the researcher. Additionally, the proposed approach requires computational resources for feature selection and classification tasks, which can be a limitation for resource-constrained systems. To be brief, this research paper presents a promising approach for feature selection and classification tasks in NIDS using an intelligent hybrid GA-PI algorithm. While there are some challenges and limitations, the proposed approach has the potential to contribute to the development of effective and efficient NIDS. 2023, Ismail Saritas. All rights reserved.

Source

International Journal of Intelligent Systems and Applications in Engineering, Vol-11, No. 7s, pp. 718-731.

Date

2023-01-01

Publisher

Ismail Saritas

Subject

Feature selection; Hybrid GA-PI algorithm; Network Intrusion Detection Systems (NIDS); Support vector machines (SVM)

Coverage

Sowmya, Christ University & Assistant professor, CMRIT, Karnatka, Bangalore, India; Mary Anita T., Christ University Bangalore, Karnatka, Bangalore, India

Rights

Restricted Access

Relation

ISSN: 21476799

Format

Online

Language

English

Type

Article

Identifier

https://www.scopus.com/inward/record.uri?eid=2-s2.0-85165153935&partnerID=40&md5=61ac7ed13d9a5cb1b861cf350fb54da6