Spam Email Classification

Utilizied Python, Pandas, and Sklearn to build a classifier which identifies spam emails with 90% accuracy

What do ‘spam’ and ‘ham’ mean?

Spam (1) emails are unsolicited or bulk messages, often commercial in nature, while ham (0) emails are legitimate, non-spam communications.

The GitHub Repo for this project can be found here!
Engineered a logistic regression model on a dataset of over 7,500 points, resulting in an email classifier that achieved 90% testing accuracy and an Area Under the ROC Curve (AUC) of 0.91.
Applied advanced feature engineering and Principal Component Analysis (PCA) to reduce dimensionality from 18 to 5 features, mitigating overfitting and ensuring 92% of legitimate (ham) emails in the test set of 1065 were correctly identified and retained in the inbox.
Enhanced model performance through GridSearch optimization across 4 hyperparameters, resulting in a 5% increase in accuracy when classifying a validation set of 1250 emails.