Communications of the ACM
Mapping Big Data
Researchers at San Diego State University are hoping geographic mapping principles will provide insights into the relationships between large sets of unstructured data.
Credit: SDSU News
Researchers at San Diego State University (SDSU) are applying geographic mapping principles to large sets of unstructured data in hopes of gaining insights into the relationships between them.
The mapping techniques being developed at SDSU's Center for Information Convergence and Strategy (CICS) focus on large sets of data that are in mutually unintelligible formats. The data can include everything from journal articles to blog posts, and CICS' mapping is meant to make the parallels and connections between them readily and quickly apparent. An example is a map the researchers made by feeding about 2 million medical journal articles into their algorithm. The map it produced enabled CICS researchers to discover that articles discussing obesity also touched on exercise, blood pressure, insulin, aging, heart rate, and school lunches, among other topics.
"How do you read a million papers? You can't. But this gives you a quick topographic overview of a new field," says SDSU professor Andre Skupin, co-director of CICS. He says CICS will serve as a resource for SDSU researchers and those in the community, but notes government agencies and public interest groups also have expressed interest.
From SDSU News
View Full Article
Abstracts Copyright © 2014 Information Inc., Bethesda, Maryland, USA
No entries found