Machine Learning, ECML- ...: ProceedingsSpringer, 2004 - Induction (Logic) |
From inside the book
Results 1-3 of 28
Page 243
... conditional entropy of all class dimensions . The conditional entropy decrease , also called conditional mutual information [ 9 ] or information gain [ 25 ] , is a very common feature selection criterion in single - objective and multi ...
... conditional entropy of all class dimensions . The conditional entropy decrease , also called conditional mutual information [ 9 ] or information gain [ 25 ] , is a very common feature selection criterion in single - objective and multi ...
Page 517
... Conditional Independence Trees In a probabilistic tree , a leaf L represents the conditional probability distri- bution p ( C | Ap ( L ) ) . If there is a representation of the conditional probability distribution over the leaf ...
... Conditional Independence Trees In a probabilistic tree , a leaf L represents the conditional probability distri- bution p ( C | Ap ( L ) ) . If there is a representation of the conditional probability distribution over the leaf ...
Page 518
Proceedings. Definition 2. A joint probability tree T is called a conditional independence tree , or simply CITree , if the local conditional independence assumption , shown in Equation 6 , is true for each leaf L. m p ( A1 ( L ) | Ap ...
Proceedings. Definition 2. A joint probability tree T is called a conditional independence tree , or simply CITree , if the local conditional independence assumption , shown in Equation 6 , is true for each leaf L. m p ( A1 ( L ) | Ap ...
Contents
Invited Papers | 1 |
RealWorld Learning with Markov Logic Networks | 17 |
Applying Support Vector Machines to Imbalanced Datasets | 39 |
Copyright | |
25 other sections not shown
Other editions - View all
Common terms and phrases
accuracy agent analysis applied approach Artificial Intelligence attributes average Bayesian network Berlin Heidelberg 2004 Boulicaut CIPER CITree classification clusters combination compute concept Conference on Machine convergence cross-validation curve Data Mining database dataset decision trees defined denotes distribution domain dynamic ECML ECTD equation error evaluation experiments extraction filter Fisher kernels function approximators graph heuristic HEXQ hidden Markov models hyperplane induction input International Conference iteration kernel learner learning algorithms linear LNAI Machine Learning matrix measure messages method naive Bayes Neural node obtained one-class SVMs optimal overfitting pairs paper parameters performance polynomial positive examples prediction probability estimates problem Q-learning random random forests ranking regression reinforcement learning representation reward rule sample SDMs Section semi-supervised learning Springer-Verlag Berlin Heidelberg SSAIR statistics supervised learning support vector machines Table task tion training data training examples training set unlabeled value function variables