Context-Aware Attack Classification

Graduate Class Project

Our work introduces an advanced detection method leveraging Long Short-Term Memory (LSTM) for identifying security vulnerabilities within the Linux kernel. Focusing on low-level context analysis and detailed per-process runtime behavior, it is meticulously designed to detect a broad spectrum of security threats while maintaining minimal impact on system performance.

PDF Analysis

Open in New Window