July 29, 2014
Natasa Miskov-Zivanov (ECE, CS) and Diana Marculescu (ECE) are among a group of researchers who have been awarded $2.4 million by the U.S. Defense Advanced Research Projects Agency (DARPA) as part of its new Big Mechanism program to create a “new way to do science and develop knowledge.” According to DARPA, “Big mechanisms are large, explanatory models of complicated systems in which interactions have important causal effects. The collection of big data is increasingly automated, but the creation of big mechanisms remains a human endeavor made increasingly difficult by the fragmentation and distribution of knowledge. To the extent that the construction of big mechanisms can be automated, it could change how science is done.”
Other members of the research group include Peter Spirtes, Professor of Philosophy at CMU; Richard Scheines, Professor of Philosophy, Machine Learning, and HCII at CMU; William Cohen, Professor of Machine Learning at CMU; and Michael Lotze, Professor of Surgery and Bioengineering at the University of Pittsburgh.
Researchers today struggle to keep up with the growing number of relevant publications and increasingly automated Big Data collection that make learning and assembling causal relationships and correlations from literature and data a laborious, manual endeavor. Why not have every publication immediately become part of a public, computer-maintained, causal explanatory Big Mechanism model of a complicated system such as cancer, brain, ecosystem or social system?
Adjunct Professor Miskov-Zivanov is leading a multidisciplinary team of researchers to develop a framework that can gather information about cancer biology from publications into Big Mechanism of cancer, update it as new papers are published or new data is collected, and make conclusions about the disease and its treatments. The team brings together experts in CAD tools for design and analysis of complex systems, causal learning, text mining, information extraction, and cancer biology.
The outcomes of the Big Mechanism program and this project are expected to significantly speed up biological research and medical practices in cancer, as well as to provide techniques and insights for developing similar frameworks in other complicated systems.