Friday, October 19, 2007

GHC: PhD Forum: Algorithms

Jo-Anne Ting started us out with her presentation on Automatic Bayesian Learning Methods. Her goal is to help systems that need to learn, learn on the fly. For example, robots or automaton cars. Her research is focused on finding the most suitable algorithm that can do this with limited memory and in real time, that doesn't get confused with with outliers.

Wei Ding gave us an overview of her research with correlation of spacial data, starting out with an example of the cholera outbreak in London in 1854, and how the outbreak was only stopped once proper analysis of the spatial data was done. She gave a good overview of how she's applied her research to analyzing problems with water wells (arsenic) in Texas & finding water on Mars. Her research results are supporting expected troubled areas in Texas - cool!

Wow - one of the folks asking a question on this paper is one of my friends from the Bay Area (Kelly), who a couple years back moved to Virginia to be a professor. I didn't know she was at this conference. Neat!

Our final presenter is Michelle L. Crane talked about her research on slicing the 3 layer architecture of UML. Her goal is to map actions to a symantec domain. I'm not familiar with UML, but I still found the discussion quite interesting.

Valerie Fenwick

. You may comment on this blog by visiting the GHC Forum.