San Francisco Datadive

On the weekend of November 4th, 2011, over 60 engaged data scientists, developers, and digital do-gooders got together to work on using data for social change at our San Francisco Datadive. Teams worked with Grameen Foundation, Mobilizing Health, and Benetech to better understand their data and build visualizations, analyses, and applications around them. Below we’ve listed some of the great work done by each team.


// Benetech

Who’s Fighting Human Rights Violations?

Benetech fights to protect human rights around the globe. Their mission is aided by a special database called MARTUS, where people can anonymously and privately upload information about human rights violations. Benetech wanted to know what kind of data was being uploaded and when so that they could have a better idea of when new outbreaks of repression were occurring and could understand how their system was being used. Click below to see how DataKind and Benetech worked to understand the data around the fight for human rights.

Overview of Results

 

// Grameen Foundation

What Makes a Good Community Knowledge Worker?

Grameen Foundation runs a Community Knowledge Worker (CKW) program where local villagers in Africa help farmers find information through cellphone searches. Grameen wanted to know how the program was working and whether certain factors made for better or worse workers. They also wanted to see if interventions like providing bikes to their workers increased their impact. See how DataKind and Grameen Foundation explored this data and discovered patterns of behavior amongst CKWs, regional differences, and built tools for assessing CKW performance.

Overview of Results | Wiki Page

 

// Mobilizing Health

How Can Villagers Get Better Health Miles From a Hospital?

Mobilizing Health brings health care to thousands of people via cellphone. They were interested in analyzing the communications between patients and doctors to better understand how patients were being served, if certain times of day were better to make requests, and whether the text of the searches held any interesting information. Click below to see how DataKind and Mobilizing Health worked with this data to identify the best doctors in the program and began building a system that could actually automatically diagnose symptoms in the absence of a doctor!

Overview of Results | Wiki Page