Detection of Neptune Attacks Using Machine Learning

This project aims to detect Neptune attacks, by analyzing network activity data using advanced machine learning techniques. The dataset contains detailed records of network activities, each labeled as normal or a Neptune attack. The approach involves preprocessing the data, addressing class imbalance with SMOTE, and utilizing a combination of RandomForest and GradientBoosting classifiers with hyperparameter tuning through GridSearchCV. The model's performance is evaluated and predictions are made on the test set to identify potential attacks.