Evaluation Methods for Machine Learning II
Papers from the AAAI Workshop
Chris Drummond, William Elazmeh, Nathalie Japkowicz, and Sofus A. Macskassy, Cochairs
The purpose of this workshop is to continue the lively and interesting debate that started last year at the AAAI 2006 workshop on evaluation methods for machine learning. The previous workshop was successful on the following points:
- It established that the current means of evaluating learning algorithms has some serious drawbacks.
- It established that there are many important properties of algorithms that should be measured, requiring more than a single evaluation metric.
- It established that algorithms must be tested under many different conditions.
- It established that the UCI data sets do not reflect the variety of domains to which algorithms are applied in practice.
At the 2007 workshop, these topics were addressed in a more specific fashion.