Too many big data initiatives are science projects that take months of effort, risk failure and require highly trained data scientists with scarce skills. According to a CSC survey, 55 percent of big data projects aren’t completed and many others fall short of their objectives.Read more...
Putting Data to Work for Public Good
Recent budget cuts have forced a heavy financial burden on many U.S. cities, some have even filed for bankruptcy, unable to pay for essential services with lower tax revenues. But Code for America, the non-profit that aims to help cities become more efficient through the power of the Internet, thinks it has a solution.
Yale alumni and the organization’s founder Jennifer Pahlka, mentions how Code for America gained inspiration from current board member Tim O’Reilly. In a keynote speech at ETech in 2008, O’Reilly made a call for hackers to “work on stuff that matters”. The group consists of tech industry leaders, city experts and “web geeks” (they don’t mind the name). Their goal is to improve how government operates through the power of the web.
This year, the San Francisco organization is hosting a four-part seminar series called “Big Data for the Public Good”, sponsored by Greenplum. The series is a discussion of how data scientists and change agents are using the growth of big data to address current challenges.
The first seminar took place on January 23, and featured speakers Michal Migurski and Eric Rodenbeck of Stamen. The focus of the seminar was data visualization. Roger Magoulas began the seminar by talking about the steps data scientists take to organize information collected from a variety of sources. He also explained that once the information is properly prepared, a new realm of possibilities open up, allowing the data to tell a compelling story.
Michal Migurski talked about Stamen’s work as a company that makes data relevant to a mass audience. A popular project completed was the Live Hurricane Maps site, in a partnership with MSNBC. He also talked about Oakland Crimespotting, an interactive, crime tracking web interface he created after becoming frustrated with Oakland’s exising crime tracker.
The next seminar will take place on March 19. Jake Porway, data scientist for the New York Times will be discussing Data Without Borders, a group that assists non-profits with resources from the data science community. The goal is to “use data in the service of humanity”
On April 3, Ward Cunningham, Nike’s Open Data for Sustainability Fellow, will discuss his collaboration with data managers. Their goals are to model product lifecycle and promote the creation of environmentally friendly products along with sound manufacturing processes.
The last seminar, titled “Data Science as a Sport”, will feature Jeremy Howard of Kaggle. Howard is both President and Chief Scientist for Kaggle, a platform that holds predictive modeling competitions. For example, the site is currently holding a $3 million competition to determine how many days a patient will spend in a hospital over the next year.
Code for America looks determined to display the advantage analytics can play for government. Given the economic climate, the group has a lot of opportunity to prove its point.