Justmephoenix/PHISHING-DETECTION-WITH-NLP ? reverse-engineered prompt

Build me a phishing detection project that uses NLP to tell the difference between spam or phishing messages and normal messages.

I want a notebook based workflow that loads the message dataset, cleans the text, turns it into useful features, trains a few machine learning models, and shows which one works best. Please include a simple way to test a new message and get a clear yes or no result, with the accuracy shown at the end. If it makes sense, use common text processing tools and a couple of basic models like SVM or random forest.

I’d also like a simple PowerBI report or dashboard for the results, so it’s easy to present the findings visually. Keep the code easy to follow, add comments where needed, and make it feel like a complete project I can open and run. If you need to check current library docs or best practices, go ahead and look them up online.