Analysis And Identification of Malware Using Machine Learning with Optimized Features selections

Electronic gadgets like laptops and smartphones are easily accessible to the people, It is attracting the interest of cyber thieves who exploit mostly through the broad availability of malicious programs as advances in technology. The increasing prevalence of malware attacks has raised the need for efficient malware detection methods. Malware is one type of program that may harm the security of a computer system. Malware is made up of unwanted, harmful opponents that attempt to hazard the security and do serious damage. As a result, malware detection and classification has become a major problem for the cybersecurity com-munity. The detection of malware consists of identifying it without knowing what it is, but malware classification involves classifying malware. Using the XGBoost algorithm to identify the importance of the feature for selection, this paper shows an effective optimized ML based approach for the identification and classification of malware. It suggests the importance of the parameter that performs best in the selection process by using RFECV techniques for the optimization of the parameters that are most effective to train the voting Ensemble of the ML models for improving the identification and classification and obtained the best accuracy 1.0 and 0.87 respectively.