Bloomberg Engineers Tackle San Francisco Civic Challenges with Bayes Impact
Originally posted on bloomberg.com
This month, engineers from Bloomberg’s Quant and Software Infrastructure teams built a toolkit that allows policy makers to study the communities of San Francisco and identify the types of investments that could have the greatest impact on a given neighborhood. The data used for the toolkit was provided by the U.S. Department of Housing and Urban Development (HUD) through the Bloomberg Philanthropies What Works Cities program.
The work was part of a 24-hour hackathon hosted by Bayes Impact where more than 300 San Francisco-based data scientists, software engineers, and designers, as well as undergraduate and graduate students at UC Berkeley and Stanford, spent their weekend using data to address civic challenges like housing and transportation. The event provided a unique opportunity for participants engage with government leaders enthusiastic about using data and technology to solve some of the biggest challenges they face.
The San Francisco toolkit aims at characterizing the livability of neighborhoods in two principal ways. First, by providing meaningful metrics for each neighborhood including data on crime, transportation and access to restaurants. Second, by investigating the relationship between those metrics and an overall satisfaction for each neighborhood.
The teams utilized bqplot, an open source library that Bloomberg recently released, to provide a unified framework for 2d visualizations with a pythonic API.
The Bayes Hackathon winning team, ‘go-bot,’ focused on using the Facebook Messenger platform to help the U.S. Department of the Interior provide State Park recommendations based on a user’s activity preference, geolocation and method of transportation.
All of the hackathon’s work is open sourced and available on GitHub.
Bloomberg is a strong advocate of using data to solve problems at the core of society. Join our next Data for Good Exchange taking place on September 25, 2016 in NYC.
