Tom Burke Portfolio

Overview:

As part of the MIT x Great Learning Applied Data Science Program, I developed a customer segmentation model to help marketing teams optimize their strategies and maximize ROI. This project built upon my AllLife Bank Customer Segmentation analysis, where I first applied clustering techniques to segment banking customers before extending my approach to a broader business application.

 

Key Contributions:
✔ Applied K-Means, K-Medoids, DMSCAN, Gaussian, and other clustering algorithms to uncover meaningful customer segments.
✔ Used Python for data preprocessing, feature engineering, and clustering analysis.
✔ Evaluated cluster effectiveness using Silhouette Score and Elbow Method to determine the optimal number of segments.
✔ Translated complex data into actionable insights for marketing teams, enabling them to target customer groups more effectively.

 

Results & Impact:

• Successfully delivered the capstone project on time, meeting all requirements, and received an A.

• Successfully identified distinct customer segments based on behavioral and demographic data.

• Built a data-driven marketing strategy that enhances personalization and improves ROI.

• Demonstrated strong technical proficiency in Python and data science, applying real-world methodologies to solve business challenges.