Enhancing Malware Detection Using Machine Learning: A Comparative Study of Classification Algorithms

This research investigates the efficacy of various machine learning algorithms in detecting malware. Utilizing a dataset of 20,000 observations with 33 features,28 classification algorithms are applied and evaluated their performance based on accuracy, balanced accuracy, ROC AUC, F1 score, and execution time. Our results indicate that Extra Trees Classifier, XGB-Classifier, Bagging Classifier, and DecisionTreeClassifier achieved perfect accuracy, with Decision Tree Classifier also demonstrating optimal execution efficiency. This study underscores the potential of decision tree-based models for efficient and accurate malware detection and suggests further exploration with larger datasets and diverse malware types for more generalized findings.