Next: Evaluating Hypothesis and Experimental
Up: Instance-Based Learning
Previous: Lazy Versus Eager Learning
- Instance-based learners delay processing until prediction time
- they form a different local approximation for each query instance
- can model complex functions by a combination of less-complex
local approximations
- information present in the training data is never lost (is this
a benefit!!!)
- computationally expense to label new instances
- finding appropriate distance metric can be difficult and
negative impact of irrelevant attributes
Patricia Jean Riddle
Wed Jun 23 13:06:34 NZST 1999