lestercardoz11/fault-detection-for-predictive-maintenance-in-industry-4.0 ? reverse-engineered prompt

Build me a Jupyter Notebook project for fault detection in predictive maintenance, focused on bearing data from Industry 4.0. I want a clean, easy to follow workflow that loads the datasets, explores them, prepares the data, and then compares a few different approaches for spotting faults and normal behavior.

Please include implementations and results for Random Forest, a basic neural network, autoencoder based anomaly detection, LSTM autoencoder, K means, Isolation Forest, One Class SVM, Gaussian distribution based detection, and PCA. I’d like the notebooks to show the full process from data prep to evaluation, with clear charts and simple explanations so I can understand which method works best.

Make it polished and research friendly, with readable sections, helpful comments, and outputs I can use for a report or presentation. If anything needs current library guidance, look up the latest docs online.