Learning through applied problems
These projects were part of my completed postgraduate program in Data Science and Business Analytics at the University of Texas at Austin. The course supplied the scenarios, datasets, and starter notebooks. Each page explains the question, the methods I used, and what the results showed in the course data. The full HTML notebook exports are available on the individual pages.
Python foundations and exploratory analysis
What can order records tell a food delivery service about demand and customer experience?
Read the case study ↗
Statistical testing and A/B analysis
Would a redesigned news landing page help turn more visitors into subscribers?
Read the case study ↗
Supervised learning with linear regression
Which device attributes help explain the resale price of a used phone or tablet?
Read the case study ↗
Classification and decision trees
Could booking details help identify reservations at higher risk of cancellation?
Read the case study ↗
Ensemble techniques for classification
How do tree ensembles compare when classifying a historical case outcome?
Read the case study ↗
Model tuning for imbalanced data
How should a model flag generator failures when a missed failure costs more than an extra inspection?
Read the case study ↗
Unsupervised learning and clustering
Can stocks be grouped by observed price and financial characteristics without a target label?
Read the case study ↗