Machine Learning–based Malware Detection System using Static Analysis

Malware attacks have become one of the most serious threats to cybersecurity, affecting personal devices, enterprise networks, and critical infrastructure. Traditional signature-based detection approaches fail to identify newly emerging malware variants due to their reliance on predefined patterns. This research presents a machine learning–based malware detection system utilizing static analysis techniques. The system extracts discriminative features from Portable Executable (PE) files, including header attributes, imported functions, and entropy measures, and applies supervised learning algorithms such as Random Forest, Support Vector Machine, and Decision Tree classifiers. Experimental results demonstrate that the Random Forest classifier achieves superior accuracy, precision, and recall, effectively detecting both known and zero-day malware samples. The proposed approach offers an efficient, scalable, and adaptive solution for real-time malware detection.