Machine learning classifier that predicts whether a URL is phishing or legitimate, built with Python, pandas, and scikit-learn.
Status: In progress (started Oct 2026)
- Load and clean a labeled phishing URL dataset
- Explore features that separate phishing from legitimate URLs
- Train logistic regression and random forest models
- Evaluate with precision, recall, and F1-score
- Document results and limitations
Python · pandas · scikit-learn · Jupyter Notebook
UCI Phishing Websites Dataset (11,055 instances, 30 features; licensed CC BY 4.0)
Coming soon.