NSF grant to create new resource to accelerate 'STEM' innovations

October 18, 2011

WEST LAFAYETTE, Ind. – Four universities are creating an interactive Web platform that will provide a new way for people working on improving "STEM" research to synthesize knowledge produced through National Science Foundation investments.

The project is designed to help researchers and NSF program officers identify trends in publications and research funding. The work also aims to identify gaps in current research and funding and potential collaborators in science, technology, engineering and mathematics - or STEM - education. The effort, funded with a $3 million, four-year NSF grant, is a collaboration among research teams at Purdue University, Virginia Tech, Stanford University and Arizona State University.

The Web platform enables researchers and officials to quickly determine who is working in specific areas, their collaborators, funding sources, program officers, research papers and findings. The system visualizes complex networks of funding and research collaborations with a map created anew for each search. The network map contains clickable nodes that yield further layers of information.

"The interactive visualizations are designed to shed light on insights that may be typically hidden from a researcher or educator," said Krishna P.C. Madhavan, an assistant professor of engineering education at Purdue and the project's principal investigator. "In essence, we archive 'dark' data - the kind of data not systematically organized for search and analysis. This is really the major contribution because this sort of analysis used to take weeks to do manually, and we can do it in a matter of seconds, which is a huge leap forward."

The project is part of an NSF program called Transforming Undergraduate Education in STEM, or TUES. Researchers will create an interactive Web search and information visualization platform called Deep Insights Anytime, Anywhere (DIA2).

"This is a Web-based, knowledge-mining, interactive visualization platform," said Madhavan, who is working with co-principal investigators Niklas Elmqvist, an assistant professor of electrical and computer engineering, and Mihaela Vorvoreanu, an assistant professor of computer graphics technology.

"One primary stakeholder group for the project is researchers who are going to be submitting research proposals and who want to benefit from the proposals that have been funded in the past and the results coming out of these funded proposals. We also target NSF program officers who are making funding decisions," Madhavan said.

Teams at Virginia Tech, Arizona State and Stanford are handling different aspects of the work.

"DIA2 is led by a highly interdisciplinary team from these four universities," said Gary Bertoline, dean of the College of Technology. "This is a really strong collaboration between Purdue's colleges of Engineering and Technology."

The new Web platform evolved from a prototype called iKNEER, or Interactive Knowledge Networks for Engineering Education Research, funded by NSF's Division of Engineering Education and Centers.

"This is potentially a major new resource for the research community," said Leah Jamieson, Purdue's John A. Edwardson Dean of Engineering and Ransburg Distinguished Professor of Electrical and Computer Engineering. "The ultimate goal is to accelerate the pace of innovations and their diffusion into widespread use to benefit research, education and society."

The project includes work led by Vorvoreanu, in collaboration with colleagues at Arizona State, into "social media optimization" to ensure rapid diffusion of the innovations fostered in this project.

"Our visualizations are a powerful way to communicate data because they translate information into a graphical form that both experts and non-experts can quickly and easily understand," Elmqvist said. "We generate the visualization on the fly for each search."

The Virginia Tech team will work closely with other research partners to understand the needs of the NSF TUES community and use them to drive the design of data-mining algorithms, said principal investigator Aditya Johri, assistant professor of engineering education at Virginia Tech.

The project will benefit from previous work by Johri and co-investigator G. Alan Wang, assistant professor in business information technology, as part of the iKNEER work.

Leading the work at ASU is Ann McKenna, an associate professor in ASU's Department of Engineering in the College of Technology and Innovation. Stanford's team is led by Sheri Sheppard, a professor of mechanical engineering.

Writer: Emil Venere, 765-494-4709, venere@purdue.edu

Sources:  Krishna P. C. Madhavan, 765-496-1247, cm@purdue.edu

                  Gary Bertoline, 765-494-2552, bertoline@purdue.edu

                  Leah H. Jamieson, 765-494-5346, lhj@purdue.edu

                  Niklas Elmqvist, 765-494-0364, elm@purdue.edu

                  Mihaela Vorvoreanu, 765-496-7709, mihaela@purdue.edu