Design and implementation of artificial immune system for security of computers

Abstract

Computer security can be implemented by various methodologies the question arises whether these methodologies are sufficient enough to provide better security to the computer system soft computing and machine learning can play a significant role in providing effective security to a computer system in this thesis an architecture is being proposed to provide the security to a computer system in which the concepts of soft computing and machine learning are used the concepts of decision tree artificial neural network and genetic algorithms are successfully implemented in the framework of this architecture to identify the non self processes these proposed approaches work on the process level hence other existing approaches like antivirus tools also work within the system if any antivirus cannot identify any virus in the computer system the operating system of the computer system executes the virus and creates its processes then our proposed approaches will identify these virus processes as non self processes and user can take action against these non self processes in decision tree learning approach id algorithm is used to construct a decision tree after calculating the entropy and information gain decision trees are generated by using the set of training