Indexing tree and pruning concepts to support case-based reasoning

Case-based reasoning is thought to aid decision making because it mimics the way in which humans analyze problems. Effective recall of case situations that are similar to a current situation is a key component of this type of decision support. This paper shows how the concept of rank can be used to create compact indexing trees for recall of similar cases. We further consider how pruning can be applied to improving compactness and methods for measuring the reliability of the result.