Showing papers by "Sameep Mehta published in 2002"

Mining of Complex Evolutionary Phenomena

Abstract: Numerical simulations are replacing traditional experiments in gaining insights into complex physical phenomena. Given recent advances in computer hardware and numerical methods, it is now possible to simulate physical phenomena at very fine temporal and spatial resolutions. Analyzing the datasets produced by such simulations is extremely challenging, given the enormous sizes of the datasets involved, and the fact that often the information that is to be gleaned is like searching for a needle in a haystack. In order to make efficient progress, analyzing such data must advance from current techniques that visualize static images of the data, to the automated mining, identification and subsequent visualization of the important features in the data, a challenging task. We report progress on a unified framework that addresses this critical challenge for two science drivers. In both our science drivers – solid and fluid systems – there are hidden hierarchies of features and the components that characterize them (e.g. shapes). Here, we offer a systematic way to detect such features, characterize and track them, and formulate hypotheses about their evolution – an important step in extracting vital information from such complex systems. The search is made more difficult due to the multiscale nature of the phenomenon being studied. Some features persist for awhile and manifest in global macroscopic effects. Other short-lived transients arise from the occurrence of certain precursor events that are important for understanding the evolutionary behavior of features.