Network Intrusion Detection with Feature Elimination and Selection Using Deep Learning

In the contemporary digital landscape, marked by an ever-increasing number of individuals seeking to exploit vulnerabilities in systems and cause immense damage to users, it becomes imperative for users to safeguard their systems, detect threats correctly, and carry out an appropriate response. Implementing Intrusion Detection Systems (IDS) is vital for monitoring network traffic, spotting threats, and swiftly alerting users to suspicious activities. It is categorized into two parts: Network-based (NIDS) and Host-based (HIDS). NIDS is strategically placed and monitors data flow across devices. A major NIDS challenge is the prevalence of false positives and, more critically, false negatives, potentially allowing undetected threats to cause significant harm. To improve intrusion detection, a new approach using Feature Selection and Deep Learning framework is used for Network Intrusion Detection System. The study analyzes metrics like accuracy, precision, F1 score, and confusion matrix.