Next: Why or When Accuracy?
Up: Evaluating Hypothesis and Experimental
Previous: Comparing Algorithms w/ Small
- Given two learning algorithms and and datasets from
several domains, which algorithm will produce more accurate
classifiers when trained on examples from new domains?
- This is perhaps the most fundamental and difficult question in
machine learning.
- We need to combined the results from several answers to question
8, where each answer has an associated uncertainty.
Patricia Riddle
Fri May 15 13:00:36 NZST 1998