An Objective Function Based on Bayesian Likelihoods of Necessity and Sufficiency For Concept Learning in the Absence of Labeled Counter-Examples.
Andrew SkabarPublished in: IC-AI (2004)
Keyphrases
- concept learning
- counter examples
- objective function
- posterior probability
- maximum likelihood
- inductive logic programming
- bayesian networks
- feature construction
- constructive induction
- optimal solution
- negative examples
- concept detection
- lower bound
- inductive learning
- training data
- supervised learning
- logic programs
- data sets
- relational databases
- domain knowledge