Case based reasoning
a problem solving strategy using existing cases
- to automate ‘knowledge reuse’
- assume previous cases have been correctly dealt with
- cases might have been addressed by humans
associate with a case a set of feature-value pairs
- together form a unique index for the case
- possibly weight features with importance score
use existing case database to help with new cases
- calculate index of new case
- find some number of the ‘closest’ cases
- use these to help treat new case