Skip to content
Open access

Explainable AI for Malware Detection: Feature Importance Analysis Using SHAP and Machine Learning

2026 · International Journal of All Research Education & Scientific Methods · 0 citations

TL;DR

A machine learning-based malware detection framework using features extracted from Windows Portable Executable (PE) files is presented and the SHAP-based feature importance approach helps identify the most influential features responsible for malware detection while also reducing model complexity and computational cost.

Abstract

Malware attacks have increased significantly recently, posing serious threats to computer systems, network infrastructures, and critical digital services. Traditional malware detection techniques mainly rely on signature based approaches, which are often ineffective in detecting newly emerging, polymorphic, and previously unseen malware variants. To address this limitation, machine learning techniques have been widely adopted to enhance malware detection by identifying hidden patterns and structural characteristics in executable files. This paper presents a machine learning-based malware detection framework using features extracted from Windows Portable Executable (PE) files. The dataset, stored in a structured format such as a malware.csv file, is preprocessed and analyzed using several machine learning classifiers, including Support Vector Machine (SVM), Random Forest (RF), Decision Tree (DT), Naïve Bayes (NB), Logistic Regression (LR), and K-Nearest Neighbors (KNN). To improve model transparency and interpretability, Shapley Additive Explanations (SHAP) are employed to analyze the contribution of individual features toward classification decisions. The SHAP-based feature importance approach helps identify the most influential features responsible for malware detection while also reducing model complexity and computational cost. Experimental results demonstrate that integrating SHAP with machine learning classifiers enhances detection accuracy and provides meaningful explanations for model predictions. The proposed approach highlights the effectiveness of explainable machine learning techniques in developing accurate, interpretable, and efficient malware detection systems for modern cybersecurity applications.

Read PDF

Similar papers

Open access Jul 2026

AI-Based Adaptive Malware Detection Using Portable Executable (PE) Header Analysis and Hybrid Machine Learning

Experimental results demonstrate that the proposed approach achieves high classification performance while improving transparency in malware detection decisions, making it suitable for practical cybersecurity applications.

V. Padmapriya, S Uma, S Sumathi et al. · 0 citations
Open access Aug 2026

Machine Learning-Based Malware Detection: A Comparative Study of Random Forest, Decision Tree, KNN, and Linear SVM

Random Forest achieves the highest performance with a test accuracy of 96.3%, F1-score of 0.947, and AUC of 0.993, establishing it as the optimal algorithm for static malware detection tasks and establishing it as the optimal algorithm for static malware detection tasks.

Umesh Balami, Ganesh Gautam, Gajendra Sharma · 0 citations
Open access Jul 2026

A Robust Deep Learning Framework for Intelligent Malware Detection

Malware has become one of the biggest threats to computer systems and digital networks, affecting individuals, businesses, and government organizations. Traditional malware detection methods mainly depend on signatures and predefined rules, making them less effective against newly developed and constantly evolving attacks. As cybercriminals continue to use advanced techniques such as code obfuscation and polymorphism, there is a growing need for smarter and faster detection methods. This project proposes a robust malware detection system using deep learning to identify both known and unknown malware with improved accuracy. The model automatically learns meaningful patterns from malware data without relying heavily on manual feature extraction. By analyzing the behavior and characteristics of malicious files, the proposed system can classify malware efficiently while reducing false alarms. The approach improves detection speed, enhances security, and supports real-time threat analysis. This makes it a reliable solution for protecting modern computer systems against rapidly changing cyber threats.

S.Srikar, G.Rajini · 0 citations
Open access 2026

Large-Scale Static Malware Detection Using Classical Machine Learning Models: An Evaluation on the EMBER Dataset

The results suggest that well-designed classical ensemble approaches are still quite competitive for scalable and interpretable static malware detection even with the rising usage of more and more powerful machine learning architectures.

A. Fauzan, Tito Pinandita, Aulia Desy Nur Utomo · 0 citations
Open access Aug 2026

Enhancing malware detection reliability in non-executable files using confidence score prediction

Malware attacks targeting widely used non-executable formats, namely Microsoft Office and PDF files, have become a prevalent threat. These files, which encompass a broad spectrum of data types are classified as complex files. Existing malware detection models currently lack transparency, providing only binary labels without confidence scores. Incorporating confidence score enhances interpretability and detection accuracy. This article proposes a learning-based malware detection approach including two complementary parts. The first part involves the development of binary classifiers, on an enriched dataset of related files, with an extended feature set to achieve high accuracy. The second methodology employs regression models to ascribe a confidence score to each sample. A reliability score is assigned to various antiviruses to accurately label samples with confidence scores. By completion of the detection process, a pair consisting of x and y is provided, where x is the binary classifier output and y is the regressor output, showing the confidence score. Our findings demonstrate an enhancement with respect to the current state-of-the-art malware detection classifiers, with accuracy improvements of approximately 2.44%\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$2.44\%$$\end{document} for PDF files and 2.27%\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$2.27\%$$\end{document} for MS Office. Using confidence score along with binary classification boosts detection accuracy to 99.74%\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$99.74\%$$\end{document} for PDFs and 99.77%\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$99.77\%$$\end{document} for Office files.

Rasoul Rezvani-Jalal, Morteza Zakeri, S. Parsa et al. · 0 citations
Open access Aug 2026

Intelligent malware detection on Android smartphones via a hybrid approach using gradient boosting and convolutional neural network

Evaluation using metrics such as accuracy, precision, F1 score, and false positive rate indicates that CNN-GBM outperforms existing deep learning models, and enhancements stem from the effective integration of CNN feature extraction with GBM’s boosting capabilities.

C. Chimeleze, Norziana Jamil, Z. M. Zain et al. · 0 citations