Logan Cooper's Portfolio

About Me

I'm a data scientist and a software developer with a diverse set of experience that runs the gamut from building NLP APIs in Python, to doing research at the intersection of economics and machine learning, to building full-stack JavaScript web apps. If my experience sounds like a good match for you, feel free to get in touch!

Check out my resume!

Projects

The decision boundary of a machine learning classification algorithm

Modle: The ML Guessing Game

A game where you are shown the decision boundary of a classification algorithm and have to guess which algorithm did the classification.

Go to Modle

A complex graph with hundreds of nodes and thousands of edges between them

Debunked! -- Modelling the LOCO Conspiracy Corpus

A text analysis of thousands of conspiracy theory articles in the LOCO Conspiracy Corpus with the goal of identifying how different conspiracy theories are linked together.

Go to Debunked!

A Jupyter notebook with a data table below it

Buck 65 Lyrics

An analysis of rapper Buck 65's vocabulary inspired by Matt Daniels' Hip-Hop Vocabulary Project

Go to Buck 65 Lyrics

A scatter plot of orange points with a red regression line running through it

Regression API

A Flask API which lets you uplaod two variables in CSV or TSV form and run a linear regression on them.

Go to Regression API

Starry Night

A JavaScript copy of the After Dark Starry Night screensaver.

Go to Starry Night

The Price of War

The Price of War is an interactive map built with d3 and Datamaps that illustrates the role of the U.S. in the global arms trade from 2001 to 2016. It uses data from the Stockholm International Peace Research Institute's arms Transfer Database.

Go to Price of War

Research

Examining the Blavatskyy Hypothesis: Obesity and Corruption in the Former USSR

In-progress paper with Professor Charlie Becker rigorously examining the results of Pavlo Blavatskyy's paper "Obesity of politicians and corruption in post-Soviet countries".

Read the Paper See the Code

PollyDarton -- NLP-Driven Data Curation for Polymer Nanocomposite Data Resources

Research applying NLP techniques to the classification of sentences in materials science papers with the eventual aim of helping to automate the discovery of polymer recipes.

Read the Paper See the Code

Testing Brueckner & Rosenthal: Did America's Downtowns Get Rich?

A paper testing a hypothesis put forward in a 2009 paper by Brueckner and Rosenthal that posits that American downtown areas will become wealthier by 2020.

Read the Paper See the Code