JayantUppal/Intrusion-Detection-System — reverse-engineered prompt

Build me a simple intrusion detection machine learning project in a Jupyter notebook. It should use the KDD Cup 1999 network traffic dataset and classify each connection as normal or malicious.

Please make the notebook easy to follow for someone learning this. Start by loading the dataset, cleaning it, converting text categories into numbers, removing unnecessary or duplicate data where appropriate, and preparing it for training. Then include some basic exploratory charts and summaries so I can understand the data and attack labels.

Train three models, Decision Tree, Gaussian Naive Bayes, and XGBoost, then compare their accuracy and false positives. After that, create an ensemble prediction using majority voting across the three models and show whether it improves the results. Include clear outputs, confusion matrices, and a short explanation of what each step is doing.

Also include a requirements file and simple setup instructions so I can run it locally with Jupyter. Look up current docs online if needed.