Automated Refinement of First-Order Horn-Clause Domain Theories
Machine Learning
Multistrategy Theory Revision: Induction and Abductionin INTHELEX
Machine Learning - Special issue on multistrategy learning
Learning Logical Definitions from Relations
Machine Learning
Refining Complete Hypotheses in ILP
ILP '99 Proceedings of the 9th International Workshop on Inductive Logic Programming
Inference for the Generalization Error
Machine Learning
A study of cross-validation and bootstrap for accuracy estimation and model selection
IJCAI'95 Proceedings of the 14th international joint conference on Artificial intelligence - Volume 2
Revising first-order logic theories from examples through stochastic local search
ILP'07 Proceedings of the 17th international conference on Inductive logic programming
ECML'05 Proceedings of the 16th European conference on Machine Learning
Online structure learning for Markov logic networks
ECML PKDD'11 Proceedings of the 2011 European conference on Machine learning and knowledge discovery in databases - Volume Part II
| Hi-index | 0.01 |
Theory revision systems are designed to improve the accuracy of an initial theory, producing more accurate and comprehensible theories than purely inductive methods. Such systems search for points where examples are misclassified and modify them using revision operators. This includes trying to add antecedents in clauses usually generated in a top-down approach, considering all the literals of the knowledge base. This leads to a huge search space which dominates the cost of the revision process. ILP Mode Directed Inverse Entailment systems restrict the search for antecedents to the literals of the bottom clause. In this work the bottom clause and modes declarations are introduced to improve the efficiency of theory revision antecedent addition. Experimental results compared to FORTE revision system show that the runtime of the revision process is on average three orders of magnitude faster, and generate more comprehensible theories without decreasing the accuracy. Moreover, the proposed theory revision approach significantly improves predictive accuracy over theories generated by Aleph system.