Evolutionary-Based Deep Stacked Autoencoder for Intrusion Detection in a Cloud-Based Cyber-Physical System

As cyberattacks develop in volume and complexity, machine learning (ML) was extremely implemented for managing several cybersecurity attacks malicious performance. The cyber-physical systems (CPSs) combined the calculation with physical procedures. An embedded computer network monitor control procedure, commonly feedback loops whereas procedures affect calculations conversely, at same time, ML approaches were vulnerable to data pollution attacks. Improving security attaining robustness of determined schemes critical problems growth CPS. This study develops a new Stochastic Fractal Search Algorithm Deep Learning Driven Intrusion Detection system (SFSA-DLIDS) cloud-based CPS environment. presented SFSA-DLIDS technique majorly focuses on recognition classification intrusions accomplishing from approach primarily performs min-max normalization convert input compatible format. In order reduce curse dimensionality, SFSA is applied select subset features. Furthermore, chicken swarm optimization (CSO) deep stacked auto encoder (DSAE) utilized identification intrusions. design CSO algorithm parameter DSAE model thereby enhances classifier results. experimental validation tested using series experiments. results depict promising performance over recent models.