Use Transformer/BERT for feature extraction and RNN/CNN/LSTM for classifying malicious HTTP requests, with model optimization using SageMaker pipeline

Requirements: Feature Extraction: Use Transformer or BERT for extracting relevant features from HTTP request data. Classification Models: Test different downstream models (RNN, CNN, LSTM) for classifying malicious HTTP requests based on extracted features. Model Optimization: Use AWS SageMaker pipeline to automate the training, evaluation, and tuning of the models. Use model tuning to identify the best hyperparameters and model architecture. Dataset: Train the models using the HTTPDataset 2010, which contains labeled HTTP traffic data for training the models. Performance Metrics: Identify the best-performing model based on accuracy, precision, recall, and F1 score. Deployment: Deploy the best model to production for real-time malicious HTTP traffic detection.