Zero Day Attack Detection and Simulation through Deep Learning Techniques
Aalap Arun, Anjaly S. Nair, A. G. Sreedevi
Abstract
Combating zero-day attacks is essential in the age of ongoing cyber threats. For simulating and identifying these dangers, our research uses Long Short-Term Memory (LSTM) algorithms, which are skilled at collecting temporal data correlations. This novel method signals a paradigm shift in cybersecurity. The project begins by developing a complex framework to model intricate zero-day attacks that imitate previously undiscovered vulnerabilities. Our detection method uses recurrent neural networks and sophisticated gating techniques, and it is LSTM-based. Its capacity to recognize novel assault patterns and pinpoint minute departures from the usual is demonstrated by rigorous testing. The result is a powerful zero-day attack detection system that improves accuracy and responsiveness. This reduces the potential damage that unknown vulnerabilities could inflict.