Dynamic Feature Optimization with Moth-Flame Optimization Algorithm for Intrusion Detection using Deep Learning

  • Unique Paper ID: 166821
  • Volume: 11
  • Issue: 2
  • PageNo: 2105-2112
  • Abstract:
  • The issue of network security has drawn increasing attention as the Internet has grown rapidly. An important area of study in network security is the detection of anomalous behaviour in networks. Intrusion Detection Systems (IDSs) are used to analyse network data and identify unusual network behaviour. IDSs can be classified into two main categories: signature-based and anomaly-based detection systems. Signature-based detection systems, such as Snort intrusion detection systems, create libraries of signatures for known malicious behaviours and compare network data against these signatures to detect intrusions. This paper proposes a feature optimization-based intrusion detection approach. For feature optimization, the Moth-Flame Optimization (MFO) algorithm is employed. The feature optimization algorithm reduces complex features and improves the detection process. For classification, a Convolutional Neural Network (CNN) is used, which enhances the system's detection capacity. The proposed algorithm was tested using MATLAB2018R software with the KDDCUP2003 dataset. It was compared with existing algorithms such as CNN and CNN-GUR. The performance analysis suggests that the proposed algorithm is more efficient than the existing algorithms.

Cite This Article

  • ISSN: 2349-6002
  • Volume: 11
  • Issue: 2
  • PageNo: 2105-2112

Dynamic Feature Optimization with Moth-Flame Optimization Algorithm for Intrusion Detection using Deep Learning

Related Articles