An Efficient Feature Extraction Method For Static Malware Analysis Using PE Header Files

Abstract

Detecting malware is crucial for safeguarding various devices, ranging from per sonal computers to large-scale systems,because computer systems continue to face serious security concerns from an increasing number of malware occurrences. Static analysis offers the ability to extract multiple file characteristics across var ious categories of information, eliminating the expenses and risks associated with dynamic analysis. By leveraging PE header information in machine learning classi fiers, an efficient feature extraction method can be developed to minimize the time required for feature extraction and therefore improve the analysis process. The objective is to enhance extraction time while maintaining a reasonable balance with other parameters, such as execution time, accuracy, and f measure.