What is spam-email-classifier?

Spam Email Classifier is a beginner-level machine learning project that takes a text message or email and predicts whether it is spam or a legitimate message, called "ham". It was built as a portfolio project for a first-year engineering student and includes a live web demo you can try in a browser. The way it works: the model was trained on the SMS Spam Collection dataset from Kaggle, which contains over 5,000 messages already labeled as spam or not spam. Before training, text is cleaned up by converting everything to lowercase, removing punctuation, and stripping common filler words. The cleaned text is then converted into numbers using a technique called TF-IDF, which stands for Term Frequency-Inverse Document Frequency, a way of scoring how important each word is in a message relative to all messages. A Multinomial Naive Bayes algorithm then uses those scores to classify new messages. The trained model and vectorizer are saved as files so the app can load them instantly without retraining each time. The front end is a simple web interface built with Flask, a lightweight Python web framework. You type or paste a message, submit it, and the app returns a spam or not-spam verdict. Setting it up locally involves installing dependencies, running a training script to generate the model files, and then starting the Flask server. The project is written entirely in Python, using Pandas and NumPy for data handling, Scikit-learn for the machine learning components, and NLTK for text preprocessing.