Optimizing CTR: Looking Back to Look Forwards

A retrospective analysis of web event data won't tell you how to maximize your click-through, but it can provide a critical first step in determining how to optimize your user engagement. I explore how to turn descriptive statistics into hypotheses, using web event data from a ficticious financial news media company.

Read More

Yelp me Out: Finding Similar Restaurant Reviewers

How can online review sites such as Yelp identify similar people among their millions of users, and use this information to enhance their platform?

Read More

Fly the Frustrating Skies: Sentiment Analysis of the U.S. Airline Industry

How does the performance of various supervised and unsupervised machine learning methods compare when tasked with identifying negative, complaint-ridden tweets sent to U.S. airlines?

Read More

Interactive School Closure Map

This map shows which counties were hit the hardest by public school closures 2009-2015. Hover over the counties to explore the closure rates.

Read More

Predicting School Closure

Can we identify schools at risk for closure by performance and other characteristics?

Read More

Oscar Bait: A Scientific Investigation

Do films with "Oscar Bait" characteristics (e.g. large, lavish period drama's) actually tally up more Oscar nominations? Using data scraped from boxofficemojo.com, I set out to investigate.

Read More

The Double Edged Sword of Data

A common pitfall of data-driven decision making is to try to replace intuition and common sense with scatter plots and p-Values. Here, I elaborate with an example from an analysis of NYC MTA turnstile data.

Read More