A Supervised Machine Learning Algorithm for Detecting Malware

The proliferation of malware is a threat to our computing system and its security.That is why the need for malware detection using machine learning arises.This work was motivated by the limitation of [1], [2] in 'Malware Detection Module using Machine Learning Algorithms.The objective of this research is to develop a security system for the detection of malware using supervised machine learning algorithms and also to carried out performance evaluation.Feature selection (Filter method) was used to reduce 100,000 columns and 35 rows of features to 20 features, then three classifier algorithms were employed which are K-Nearest Neighbor, Decision Tree and Random Forest.The classifiers are trained and tested using the dataset(malware.csv)gotten from Malware Detection Kaggle.The results of the algorithms (K-Nearest Neighbor, Decision Tree and Random Forest) are respectively 96.53%,97.79%and 99.90%.The results were also compared with other researchers [3] work that used the same three classifiers, the results of Maqsood 2020 for Random Forest, Decision tree and K nearest neighbor are respectively 96.39%, 100%(overfit) and 99.4%, while the results of Sarang et al 2013 for Random Forest, Decision tree and K nearest neighbor are respectively 99.57%, 99.23%, and 99.06%.It indicates that Random Forest is most effective out of the three classifiers algorithm for malware detection using machine learning, moreover, the study performed can be useful as a base for further research in the field of malware analysis with machine learning methods.