Journal of Engineering and Engineering Technology - JEET |BIO-INSPIRED FEATURE EXTRACTION TECHNIQUES FOR INTRUSION DETECTION SYSTEMS

Paper Details

Title :BIO-INSPIRED FEATURE EXTRACTION TECHNIQUES FOR INTRUSION DETECTION SYSTEMS
Author(s) : Folasade Adeyemi (1), Samuel Adebayo Oluwadare (2) and Boniface Kayode Alese (3)
DOI:
Abstract:

Computer networks and systems are protected against unauthorised access and harmful activity using intrusion detection systems (IDSs). IDS are very important to the security architecture of computer networks. They are at the forefront of detecting harmful activities and unauthorized access to the network. IDS usually manages large amounts of data traffic containing redundant and irrelevant features. Due to its potential to increase detection accuracy, decrease computing burden, and increase system scalability, feature selection has drawn significant attention as a crucial pre-processing stage in IDS design and was shown to significantly affect the classifiers' performance. To create appropriate detection algorithms, the set of features thought to be the most useful properties are retrieved during feature selection. An in-depth analysis of feature selection methods within the framework of IDS is presented in this paper, with a focus on bio-inspired strategies. Bio-inspired strategies offer creative answers to challenging issues by taking their cues from natural systems and processes. This paper explores and evaluates the effectiveness of bio-inspired algorithms for feature selection in intrusion detection systems (IDS) to enhance detection performance. This framework provides a thorough understanding of the current landscape of feature selection for IDS with bio-inspired techniques by critically evaluating the state-of-the-art research, challenges, and opportunities facilitating advancements in this important area of cyber security.

Pages

Contact us

Email us

Call Us