Phishing Detection Across Email, URL, and SMS: Reproducible Machine Learning Experiments

This repository contains the code package and supporting materials for reproducible machine learning experiments on phishing detection across email, URLs, and SMS messages. The experiments evaluate two baseline classifiers, Logistic Regression and Linear Support Vector Machine (SVM), using TF-IDF text features. The evaluation includes accuracy, precision, recall, F1-score, and average precision. The experimental pipeline combines samples from three phishing-related data sources, performs text preprocessing and label mapping, and applies a group-based train-test split to reduce duplicate-text leakage between training and testing data. The repository supports reproducibility through documented preprocessing, fixed random seeds, dataset auditing, model evaluation, and generated performance metrics. This upload is intended to support an academic research project on phishing detection using machine learning. Original third-party datasets are not redistributed unless their licenses permit redistribution.