A real-time dashboard for quickly visualizing tends in currency fluctuations for major currencies over the past month. Developed end-to-end to display metrics in real-time. Extracts metrics from the ECB currency API, transforms data using Python's Pandas, visualizes sparklines using d3. Served using nginx and gunicorn via the Flask micro-framework.
A dynamic dashboard for visualizing a movie rental company's key business metrics. Developed end-to-end to display metrics in real-time.Extracts metrics from a MySQL database, transforms data using Python, visualizes charts and graphs using d3. Served using nginx and gunicorn via the Flask micro-framework.
A dynamic visualization of the unemployment rates in OECD countries over the past 15 years.
A dynamic visualization of presidential voter turnout by country over time.
A Django-based application that leverages the Google Maps API to determine wheter a clicked-on location has an address or not.
A dynamic visualization of the medal count for the Olympic Games by gender over time.
A Pig and Bash tool to find ith-degree connections for a list of 1st-degree connections between nodes. This tool uses a Pig abstraction layer to create Map Reduce jobs that allow for scaling to a very large number of connections.
A dynamic, visual representation of the Siple windchill formula.
A dynamic representation of Periodic Table electronegativities.
Project undertaken at Twitter to determine the offline sales impact of social networking platforms using a Geo-Based Methodology. Presented at QCon and Grace Hopper.
A simple, dynamic data visualization using Python's Pandas and Bokeh libraries.
I explore the following research question: "What determines the levels offunding that an NSF or NASA-funded project receives?". I define high levels of researchfunding as the upper decile for the original, raw funding variable. I then use text analytics to determine the words in the abstract that were correlatedwith high levels of funding. Using CART, logistic regression, and Random Forest models,I am able to predict whether a research project is likely to have received high levels of fundingwith around 90% accuracy.
Used R to predict the development of the energy industry usinga patent dataset. Determined whether certain technologies had reached a saturation point.
Created an online, interactive Tableau dashboard for the Inter-AmericanDevelopment Bank that visualized development indicators across various racesand ethnicities in Latin America. Scraped national census data to createdatabase that populated the dashboard.
Created a real-time, dynamically updating social media dashboard that tracks a Twitter account's key use statistics.