This is a project that is used to predict network intruions. A system for detecting and classifying network attacks. The proposed Network Intrusion Detection System employs a hybrid approach, leveraging a Random Forest model to determine whether a network packet is benign or malicious. If classified as malicious, the attack is further categorized into one of four types- DoS, Fuzzers, Exploits, or Reconnaissance using a 1D Convolutional Neural Network (CNN). The system also records key details such as the attack sources, destinations, and packet attributes in a database. Additionally, it generates comprehensive reports from the collected data, enabling users to analyze attack patterns and trends effectively.