Artificial Neural Network Based Approach for Identification of Operating System Processes

Abstract

A computer system can be secured by using various methods like firewalls, anti-virus tools, network security tools, malware removal tools, monitoring tools etc. These tools and applications are being by most of the computer users. These computer security tools need to be updated and monitored regularly by the user. If any computer users fail to update the security tools, then the computer system may be infected by virus or may be attacked. Through this paper a learning system is being proposed to provide security by identify the operating system process as Self and Non-Self. Concepts of Artificial Neural Network (ANN) Learning have been used for the identification of processes. Initially, an Artificial Neural Network is created by using processes parameters with random weights. These weights are updated by using Gradient Descent Algorithm for various training examples, and then this Artificial Neural Network is tested with test data examples. It has been observed that the Artificial Neural Network Learning provides a better approach for identifying Self and Non-Self process and provides a better security.

Keywords: Self and Non Self Process, Machine Learning, Artificial Neural Network, Gradient Descent, Perceptron

View PDF