NETWORK BASED INTRUSION DETECTION SYSTEM USING DEEP LEARNING

Abstract

With the rapid increase in cyber threats, traditional intrusion detection systems (IDS) struggle to keep up withsophisticated attacks. This project aims to develop an Advanced Network Intrusion Detection System (NIDS)using Deep Learning techniques to detect and classify network intrusions effectively. The system processes realtime network traffic and classifies it as normal or malicious using deep learning models such as ML models. Thedataset is preprocessed using feature engineering techniques like One-Hot Encoding and Min-Max Scaling toimprove accuracy. The trained model is deployed in a Flask-based web application that continuously monitorsnetwork activity and alerts administrators about potential threats. Unlike traditional signature-based IDS, thissystem can detect zero-day attacks by learning patterns from previous intrusions. By comparing multiple deeplearning architectures, we aim to achieve high accuracy, precision, and recall in intrusion detection. The proposedsystem enhances network security and helps organizations prevent unauthorized access and data breacheseffectively