4,270,000 tweets, blog posts, and news articles combined. Using a custom katz-backoff model process to rapidly search text and return predicted words and probabilities.

Capstone in the JHU Data Science Specialization

Report (with code) on the relationship between inflation and unemployment.

Slideshow on the relationship between inflation and unemployment.

Slideshow of healthcare.gov comment data analysis

Using the World Bank's Poverty and Inequality Data from 1975 to 2015 I analyzed the dataset, trimmed it down as carefully as possible, and made a few key visualizations.

Reproducible Download and some exploration, does not yet include model building.

Why should you use my NBACompare Shiny app?

Feature Selection, Data Cleaning, Model comparisons, parallel computing.

Using techniques from JHU Coursera Machine Learning W4 video, combining predictors. I followed along, but made a reproducible version with corrections for updates to predict(). Weirdly, the GAM ensemble model didn't beat the basic GLM model for the validation set.

Attempting Plot_ly with mtcars data

Final Project for Regression Modeling coursera course.

Using population data, coordinates, and leaflet() to make an interactive map of the smallest cities in the USA

I made this for fun and to test combinations of ggplot2 and aesthetic themes.

Designing four categories and then testing regular season boxscores from basketball-reference. Westbrook and Isaiah Thomas tied for MVP. I'm personally picking Westbrook to win, but it didn't feel right to design a tie-breaker after having seen the data.

Orange Juice vs Abscorbic Acid as a catalyst for cellular tooth growth in guinea pigs. For low doses of Vitamin C the supplement used has a major effect on too growth, but beyond 2mg/day the supplement stops mattering.

Used rexp() to simulate data and explore means and variances around lambda = 0.2

An analysis of NOAA data to identify the deadliest and damaging weather events.