Email Classification & Spam Detection Engine
An end-to-end Machine Learning pipeline built to clean, vectorize, and classify text data with high accuracy using automated preprocessing and evaluated classification algorithms.
- Engineered an automated data preprocessing pipeline in Python using Pandas & NumPy to clean and tokenize text.
- Trained and evaluated classification models using Scikit-Learn, achieving 90%+ classification accuracy.
- Performed feature extraction and TF-IDF vectorization to detect predictive text patterns.