Virus detection using data mining techinques

Abstract

Malicious executables are computer programs, which may cause damages or inconveniences for computer users when they are executed. Virus is one of the major kinds of malicious programs, which attach themselves to others and usually get executed before the host programs. They can be easily planted into computer systems by hackers, or simply down loaded and executed by naive users while they are browsing the Web or reading e-mails. They often damage its host computer system, such as destroying data and spoiling system software when they are executed. Thus, to detect computer viruses before they get executed is a very important issue. Current detection methods are mainly based on pattern scanning algorithms. However, they are unable to detect unknown viruses. An automatic heuristic method to detect unknown computer virus based on data mining techniques, namely decision tree and naive Bayesian network algorithms, is proposed and experiments are carried to evaluate the effectiveness the proposed approach.