Network Activity Classification Using Random Forest for SYN Flood Detection

For a machine learning challenge focused on network activity classification, I employed a Random Forest Classifier to distinguish between normal network behavior and Neptune (SYN flood) attacks. After preprocessing the training data (train.csv) by handling missing values, applying one-hot encoding, and addressing data imbalance using SMOTE, I tuned the model's hyperparameters via GridSearchCV to optimize F1 score. Training on resampled and scaled data, I evaluated model performance using validation metrics and prepared predictions for the test data (test.csv). The resulting submission file (submission.csv) includes predictions formatted to match competition requirements, aiming to accurately classify network activities based on provided features