The main goal of this project is to solve the problem of network intrusion detection system, we have used deep learning and traditional machine learning techniques to enhance intrusion detection. Intrusion detection plays a very important role in network security, which basically helps to detect malicious activities or anomaly behavior by monitoring network traffic. Or it has used traditional techniques but due to its limitations it has become necessary to integrate deep learning. In the project we used the NSL-KDD dataset, which is a widely accepted dataset for network traffic and intrusion data. Feature extraction and dimensionality reduction techniques such as PCA (Principal Component Analysis) have been used to process the data. In the algorithm we used decision trees, random forest, SVM, ANN and ensemble methods, which improved our detection accuracy. The future scope of this project is to use hybrid approach in real-time network security systems, so as to reduce false alarms along with high detection accuracy. Going forward, we can use it in distributed systems and integrate advanced deep learning models, so that even large and complex networks can be effectively secured.