Survival of the Fittest; Variable Selection on Agricultural Data from the Galapagos Islands

Masters Paper

Found key variables among 200+ candidates using Elastic Net and Forward Selection.

Hidden warnings; using topic modeling on Yelp reviews to predict health inspection violations

Class Project

A deep dive into the predictive power of Supervised Latent Dirichlet Allocation on text reviews to detect food safety concerns.

PDF Github

Yelp review text recommender system

Class Project

My first foray into text mining to build a modified recommender system with reduced sparsity.

Github

Knowledge of influenza vaccination recommendation and early vaccination uptake during the 2015–16 season

Co-author

Contributed data processing and calculation of estimates for journal article in Vaccine

Journal

Impact of Survey Device on Response to an Early Season Influenza Vaccination Survey

Lead Author/Poster Presenter

Compared responses on Mobile vs Tablet vs Computer devices to National Internet Flu Survey for CDC. Presented findings at American Association for Public Opinion Research (AAPOR) 2016 conference.

PDF

Fit for Purpose

Co-author/Poster Presenter

Evaluated suitability of web panel survey compared to traditional large-scale surveys for early-season flu vaccination estimation for CDC. Presented findings at AAPOR 2015 conference.

PDF

College Scorecard

Contributing Statistician

Programmed data processing scripts for the Department of Education’s inaugural edition of the College Scorecard.

Website