Detecting Web Attacks With End to End Deep Learning

Abstract

Websites and online applications are often targeted by hackers using attacks like SQL
injection, cross-site scripting (XSS), and denial-of-service (DoS). Traditional security systems use
fixed rules to detect these attacks, but they often miss new or advanced threats. This project uses deep
learning, a type of artificial intelligence, to detect web attacks automatically. The system learns directly
from raw data, like website request logs, without needing humans to manually create rules. We use
advanced deep learning models like CNNs and RNNs to find patterns in the data and spot harmful
activity. Our tests show that this approach is very accurate and better than older methods. This deep
learning-based system can help make websites safer and respond quickly to new types of cyber-attacks.