This project is developing novel, scalable techniques to mine massive astronomical survey databases (66 million lightcurves now available, growing to ~100 billion in a decade) for the signatures of planets around other stars. Giant astronomical surveys will soon monitor much of the sky regularly, detecting vast numbers of interesting, variable astronomical objects. These objects will range from small solar system bodies to distant quasars; in the middle of the range of scales the surveys will find planets around other stars when they transit (pass in front of) their parent stars. The objective of this proposal is to get ready for this new treasure trove.
Specifically, this project is: (1) developing new, scalable techniques to identify extra-solar planets in synoptic survey data, and (2) preserving novelty detection when the volume of data becomes so large that the scientists are "removed" from the data. The goal is to discover new extra-solar planets.
The outcome of this project will be new techniques for mining these extremely massive databases, and a preliminary survey for candidate planets. This will enable follow-up work to scour new data as they become available. The datamining tools will have impact that extends beyond the needs of astronomy. The screening of time-series data for interesting events occurs in many arenas, ranging from national defense to the tracking of commercial transactions.
