Collection of case studies and final capstone project completed during Galvanize Data Science Immersive Program. Employed modern machine-learning techniques to produce efficient, action-oriented, and presentable solutions to industry relevant problems.
Capstone: NBA Draft Model Built Gradient Boosting regression model, based on a player's college statistics, to predict their respective second-year NBA Value Over Replacement Player (VORP). Used these predications to rank college prospects in preparation for the 2017-18 NBA draft.
Churn Case Study Built Gradient Boosting model to predict churn for a ride-sharing company in San Francisco. Predicting the likelihood of churn based on user profiles in order to provide suggestions to improve user retention.
Regression Case Study Utilized ridge regression to predict the sale price of a particular piece of heavy equipment at auction based on it's usage, equipment type, and configuration.
Recommender Case Study Built recommender system utilizing surpriselib to improve upon existing recommender for video-streaming service.
Fraud Detection Case Study Created a web-based app, which pulls in streaming data from an e-commerce website. That data is then run through a Random Forest Classifier to predict whether a certain posting is fraudulent. The resulting prediction is saved to a MongoDB and displayed on the web-app.