Data Brunch and Project Pitchfest!

Data Science DC, Data Innovation DC, and District Data Labs are hosting a Data Brunch and Project Pitchfest event on Saturday April 4th from 11am - 1pm at GWU's Funger Hall. Join us!

Saturday? Yup, we are changing things up and swapping out pizza and empanadas for bagels and other brunchy foods.

Eight teams of DC data scientists have come together for a three month incubator to turn theory into practice on projects spanning healthcare, economics, the environment, and more. Learn from their experience implementing a Deep Learning network on commercially available hardware, on deploying a D3.js visualization web app using Heroku, or on building a desktop GUI with Python... plus much more! Enjoy brunch and drinks on us as we are taken from concept to production on eight data products, and then join the judges by voting for the winner!

About the teams:

SeeFish: Kathleen, Matt, and Chris are entering the Kaggle National Data Science Bowl, for which they must analyze underwater images to quantify plankton populations using Deep Learning systems.

Group D: Constance, Vinoth, and Mike are developing an algorithm to accurately categorize overall ratings by the main contributing factor (e.g. service, food quality) as identified by the text in the corresponding review using a Yelp dataset.

In collaboration with the Department of Labor:

LaborMatch: Yunan, Keegan and Ali are combining BLS data with data via social media APIs to visualize skills-based supply/demand by geographic region and predict future trends in skill development using time-series and regression analysis.

OKJK: Jim, Kathleen, and Oliver are working to form a suite of compressed economic indicators by geographic region using BLS data to improve understanding of economic health.

PPM-Data: Patrick, Mehdi, and Paul are working to predict job growth using BLS economic data and regression analysis.

Group B: Danielle, Elaine, Jing and Will are visualizing pockets of excess demand by industry, skill and resource using BLS data.

In collaboration with HIMMS:

The Synthesizers: Jamie, Kenneth and Bohan are working with HIMMS to add functionality to an open source data synthesizer to generate synthetic patient data sets with variable corruption by field.

In collaboration with ZeroCycle:

The Planeteers: Tony, Eyuel and Kate are optimizing attribution of recycling behavior to hyper local geographical regions using ZeroCycle and Census data.

If this sounds like a good time to you, RSVP on the DSDC event page or the DIDC event page. See you there!