Detection of the cyber network attack using robust random forest in a big data environment

Our civilization has entered an era of "knowledge revolution" due to the fast advancement and widespread use of IT and the Internet. Moreover, as network traffic grows and becomes more complicated, the area of cyber network attack detection faces significant new hurdles. The need for a good and efficient system for detecting attacks from a broad spectrum of network traffic serves an essential function. This article aims to provide a unique, robust random forest classifier for identifying the attack in the dataset, which focuses on distinguishing traffic inside assaults from typical big data flows (KDDs)—initializing it by defining and pre-processing the network traffic data. Then the robust random forest may be used to depict it. Cuckoo search optimization may be used to improve and optimize the network. Ultimately, to detect cyber network attacks, an extensive data test was executed. Our paper's simulations demonstrate that the approach suggested has greater detection accuracy and a higher true positive rate while simultaneously having a lower false-positive