Stacked Ensemble Method of Multi Class Malware Detection Using PE Header and Section Attributes
- Title
- Stacked Ensemble Method of Multi Class Malware Detection Using PE Header and Section Attributes
- Creator
- Kumar, Deepanshu; Poonia, Ramesh Chandra; Shanbhog, Manjula
- Description
- Malware has now become sophisticated. The type of attacks has changed, too. To identify and remove them is now a great challenge. This paper presents a machine learning model for malware detection in windows. The Malware is detected based on the static collection of features, which includes the Portable Executable (PE) Header and Section data. Several classifiers were trained on a balanced dataset, including Logistic Regression, K-Nearest Neighbour, Support Vector Machine, Multi-Layer Perceptron, XGBoost, and Stacked Ensemble. The proposed stacking method utilises SVM, MLP, and XGBoost, with XGBoost serving as the meta-learner. The model delivered the best performance when compared with all the baseline models for an accuracy of 96.25% and an AUC of 0.9978. 2025 IEEE.
- Source
- Proceedings of International Conference on Digital Innovations for Sustainable Solutions, ICDISS 2025;
- Date
- 01-01-2025
- Publisher
- Institute of Electrical and Electronics Engineers Inc.
- Subject
- classification; cybersecurity; Ensemble learning; machine learning; Malware detection; PE fields
- Coverage
- Kumar D., CHRIST University, Bengaluru, India; Poonia R.C., CHRIST University, Bengaluru, India; Shanbhog M., CHRIST University, Bengaluru, India
- Rights
- Restricted Access; Hardcopy may be available in the library
- Relation
- ISBN: 979-833155641-9;
- Format
- online
- Language
- English
- Type
- Conference paper
Collection
Citation
Kumar, Deepanshu; Poonia, Ramesh Chandra; Shanbhog, Manjula, “Stacked Ensemble Method of Multi Class Malware Detection Using PE Header and Section Attributes,” CHRIST (Deemed To Be University) Institutional Repository, accessed June 19, 2026, https://archives.christuniversity.in/items/show/25965.