Artificial Neural Network-Based Efficient Cyber Hacking Detection System Using Deep Learning Approaches

Abstract

Cyber hacking can be defined as the process of observing the incidents happening in a computer network or system and inspecting them for indications of possible incidents, which includes either violation or threats of violation in the policies of computer security, the allowable use of policies or the practices of maintaining standard security. CHS aid the network in automating the process of intrusion detection. CHPS is software that consists of all the abilities of the anomalies. In addition, it also strives to widen the possible incidents and cyber hacking methodologies with similar abilities. In the case of CHPS, it allows administrators to turn off prevention attributes in anomaly products, making them work as a cyber hacking system. Respectively, for compressing the benefits of both IPS and CHS, a novel term, cyber hacking, and prevention systems (CHPS), is used for all the further chapters to infer both CHS and IPS approaches. In this research, three algorithms, namely decision stump method (DSM), support vector machine (SVM), and artificial neural network (ANN), were used. From the results obtained, the proposed ANNAccuracy of 92.3%, MSE of 0.000119, Log Loss of 0.4288, and Mathews Coefficient of 0.9010 were proposed. The tool used is Jupyter Notebook, and the language used is Python.