Kind of interesting. Generalization can be based on hierarchies (categories) or similarities. I've thought about trying to come up with some similarity metric for meta-analysis, where studies are obviously more or less similar to each other, but not categorically related, or at least the hierarchies are extremely sparse (most categories have zero instances).
Properties of inductive reasoning
The structure and function of explanation
Subscribe to:
Post Comments (Atom)
No comments:
Post a Comment