About this document

Research Paper Supernovanet PDF by 126157006 is a document available to read on EtoBox.

The document presents SupernovaNet, an advanced deep learning model for dynamic malware detection that utilizes intrinsic API sequence features, integrating a CNN+BiLSTM+Attention architecture, ensemble learning, and SHAP-based Explainable AI. The model achieves a validation accuracy of 99.58% and an AUC of 1.00 on a dataset of 4,000 Windows PE samples, significantly outperforming existing methods. Key innovations include attention mechanisms for critical API pattern focus, robust generalization through a B

Author

126157006

Language

EN