Professor Chawla Gives Invited Talk at NASA Conference on Intelligent Data Understanding

10/23/2009:

Professor Chawla was an invited speaker at the NASA Conference on Intelligent Data Understanding (NASA CIDU). NASA CIDU conference brings together researchers in the area of Autonomous systems and systems health management, Discovery Algorithms, Data/Information Infrastructure, and Computational Methods and Frameworks. It is sponsored by the NASA's Integrated Vehicle Health Management Project, Aviation Safety Program and NASA's Applied Information Systems Research Program. The Conference was held at the NASA AMES Center, Moffett Field, CA.

Professor Chawla's talk was titled, "From Learning to Knowledge Discovery to Action in Distribution Sensitive Scenarios." The abstract for this talk is as follows: Models for knowledge discovery in the real world face the pervasive and compelling problem of irregularities in data distribution. Decisions that are optimal in expected utility can be vulnerable to failure, and value functions that reflect the discontinuities of the real world pragmatics can quickly become intractable. Surprises can happen in uncertain environments. The class distributions may not be the same (imbalanced data), with the class of interest being rare or extreme. The training and testing distributions can differ. The costs of making mistakes or benefits from making correct predictions may also not be constant and can evolve due to operational reasons. I will present some of our work on tackling such challenges in the journey from data to learning to knowledge discovery to action.