Participting in the Data Science for Social Good Incubator Program

Participting in the Data Science for Social Good Incubator Program

We’ve been selected to participate in the Data Science for Social Good summer incubator program!  It’s a really great opportunity for us to work with smart graduate students and faculty at the UW eScience Institute, while sitting side by side with three other selected projects leveraging data science for social good in urban environments.


Here’s the project description:

Assessing Community Well-being through Open Data and Social Media

With this project we will be creating neighborhood community report pages in the context of a hyperlocal, crowd-sourced community network. Our direct, “socially good” objective is to help neighborhood communities better understand the factors that impact community well-being, and how they as a neighborhood compare with other neighborhoods on these factors, to help them set the agenda for what to prioritize in promoting their well-being. A key aspect of this project is to explore novel ways to leverage diverse social media and open data sources to dynamically assess community-level well-being, in order to a) enable early identification of emerging social issues warranting a collective response, and to b) automatically identify and recommend the local community hubs best positioned to coordinate a community response. While the tools are intended to be general purpose, through the summer we will be targeting two more underserved neighborhoods in King County (the International District and the Central District). Specific project activities include:

1) Collecting and processing diverse hyperlocal social media (e.g., Twitter, Instagram, Foursquare) and open data sources (e.g., Census data, Crime data, building permits, map data) to develop community well-being measures, which may include sentiment/content analysis, social network analysis, geo-spatial analysis of hyperlocal business activity, and social media activity metrics;

2) Algorithmically integrating these metrics to develop a summary measurement model of overall “community well-being” and individual/local business “community hubbiness”;

3) Representing these metrics to end users (neighborhood community members) in neighborhood report pages, which may include overview visualizations that represent neighborhood well-being across neighborhoods; and

4) Throughout the process, actively work with one or two underserved neighborhoods (International District, Central District) to engage in community participatory design.

This works builds on prior work while our lead investigator Shelly Farnham was as Microsoft Research, exploring the use of social media to help hyperlocal communities stay informed. Related references: Facilitating Information Seeking For Hyperlocal Communities Using Social Media. Yuheng Hu, Shelly Farnham, Andrés Monroy-Hernández. In Proceedings of the SIGCHI Conference on Human Factors in Computing Systems (CHI) 2013

Neighborhood Community Well-being and Social Media, Shelly Farnham, Michal Lahav, Emma Spiro, and Andres Monroy-Hernandez. A recent, as yet unpublished paper exploring the relationship between social media usage and hyperlocal community well-being.