Hybrid Machine Learning Model for Accurate and Efficient Android Malware Prediction Using MLP-SVM
Year: 2026
Type: conference-paper
Abstract: The given research introduces a Hybrid Machine Learning Model, where Multilayer Perceptron (MLP) and Support Vector Machine (SVM) are combined to complete the task of performing accurate and efficient Android malware prediction. The proposed system is a hybrid of both the deep featurelearning potential of the MLP and the potent classification ability of SVM to develop a powerful detection system of the emerging mobile threats. Benign and malicious Android applications of very large scale were used and the hybrid model was shown to have better precision with a shorter time taken to classify the applications and overall accuracy was better than other single algorithms. The findings indicate that MLP-SVM hybrid method is an effective model of detecting complicated behavioral patterns in Android applications, which provides a secure way to improve the idea of cybersecurity in mobile ecosystems.
Language: English
Subfield: Software
Field: Computer Science
Domain: Physical Sciences
Open Access status: hybrid
OpenAlex ID: W7202204440