Analyzing Various Machine Learning Algorithms for the Classification of Malwares

Abstract

Malware, brief for Malicious Software, is increasing continuously in amounts and sophistication as our digital world continues to develop. Lately, tools for forming malware have been increasing rapidly on the internet, making it more accessible for people without expertise to create malware. Towards the end, the number of malware is growing fast. To deal with the problem, it is necessary to classify malware instantly and accurately. These malware programs play a role such as encrypting or destroying sensible data, stealing, changing or capturing main computing functions, controlling users and perform computer activity without their consent. In this paper, six different algorithms for classification like Linear Regression, Random Forest, Adaboost, Gaussian, Gradient Boosting, Decision Tree has been used for classifying a file as malicious or benign. Based on the results, the Random forest attained 99.43% accuracy and infers that it is suitable for malware prediction.