Intrusion Detection System Using Deep Belief Network With Grass Hopper Optimization Approach (DBNGOA-IDS)

Abstract

Information security is greatly aided by intrusion detection, and the essential technology is the ability to precisely identify different network threats. This research explores deep learning models for intrusion detection systems and proposes a deep learning strategy for intrusion detection utilizing Deep belief network and grasshopper (DBNGOA-IDS). Additionally, the model's performance is investigated in binary classification and multiclass classification, as well as how the number of neurons and various learning rates affect the performance of the suggested model. The performance of the present study is compared with other machine learning techniques on the benchmark data set.The experimental results demonstrate that DBNGOA-IDS performs better than typical machine learning classification methods in both binary and multiclass classification, and that it is particularly well suited for modeling a classification model with high accuracy.The DBNGOA-IDS model enhances intrusion detection accuracy and offers an innovative method to intrusion detection.