Machine Learning–Based Harmful File Detection System Using Static Analysis of Portable Executable Files

Malicious files pose a significant threat to modern computing systems by compromising data integrity, privacy, and system security.With the rapid growth of digital data exchange, traditional signature-based detection techniques are becoming insufficient to identify newly emerging and unknown threats.This paper presents a machine learning-based harmful file detection system designed to classify files as safe or malicious with improved accuracy.The proposed system analyzes various file attributes such as size, structure, permissions, and behavioral patterns to train classification models.Machine learning algorithms are employed to learn patterns from historical data and detect previously unseen malicious files effectively.The system is evaluated using standard performance metrics including accuracy, precision, recall, and F1-score.Experimental results demonstrate that the proposed approach enhances detection efficiency while reducing false positives compared to conventional methods.This work highlights the effectiveness of machine learning techniques in strengthening file security and provides a scalable solution for real-world cybersecurity applications.