Dr. Paul Gazis

December 21, 2006

"Advanced Data Mining Techniques for the Analysis of Large Space Science and Astrophysical Data Sets"


The proposed research is intended to evaluate the suitability of a several promising data mining techniques for the analysis of extremely large data sets and develop these techniques for use by the general space science community. In the process, the proposed research will address outstanding questions related to the evolution of transient events in the solar wind, the classification of near-IR and thermal emission spectra, and the identification of structure in large-scale cosmological surveys. The results of the surveys we perform to test these techniques should have immediate scientific value in their own right. The tools developed under this proposal should be immediately useful for similar surveys involving different data sets. They should also be applicable to a broad range of problems that involve large-scale surveys of multi-dimensional data or extended time series that would be difficult or impossible to perform using any other means. These tools should have immediate relevance to large-scale surveys associated with the search for extra-terrestrial intelligence.