MetaCartSign in to MyCiteSeer

Include Citations | Advanced Search | Help

Include Citations | Advanced Search | Help

  Constructive Induction in Knowledge-Based Neural Networks (1991) [16 citations — 3 self]

Download:
pdf | ps
by Geoffrey G. Towell, Mark W. Craven, Jude W. Shavlik
Machine Learning - Proceedings of the Eighth International Workshop
ftp://ftp.cs.wisc.edu/machine-learning/shavlik-group/towell.nips4.ps
Add To MetaCart

Abstract:

Artificial neural networks have proven to be a successful, general method for inductive learning from examples. However, they have not often been viewed in terms of constructive induction. We describe a method for using a knowledgebased neural network of the kind created by the Kbann algorithm as the basis of a system for constructive induction. After training, we extract two types of rules from a network: modified versions of the rules initially provided to the knowledgebased neural network, and rules which describe newly constructed features. Our experiments show that the extracted rules are more accurate, at classifying novel examples, than the trained network from which the rules are extracted. 1

Citations

2141 Learning Internal Representations by Error Propagation – Rumelhart, Hinton, et al. - 1986
317 Computer Systems that learn – Weiss, Kulikowski - 1991
120 An empirical comparison of pattern recognition, neural nets, and machine learning classification methods – Weiss, Kapouleas - 1989
83 Back propagation is sensitive to initial conditions – Kolen, Pollack - 1990
74 An Empirical Comparison of ID3 and Back-propagation – Fisher, McKusick - 1989
72 Training KnowledgeBased Neural Networks to recognize genes in DNA sequences – Noordewier, Towell, et al. - 1991
52 A comparative study of ID3 and backpropagation for English text-to-speech mapping – Dietterich, Hild, et al. - 1990
51 Interpretation of artificial neural networks: Mapping Knowledge Based Neural Networks into rules – Towell, Shawlik - 1992
39 Medical diagnostic expert system based on PDP model – Saito, Nakano - 1988
35 Refinement of Approximately Correct Domain Theories by Knowledge-Based Neural Networks – Towell, Shavlik, et al. - 1990
26 Symbolic and neural net learning algorithms: An empirical comparison – Shavlik, Mooney, et al. - 1991
24 A Network of neuron-like units that learns to perceive by generation as well as reweighting of its links – Honavar, Uhr - 1988
20 Performance comparisons between backpropagation networks and classification trees on three real-world applications – Atlas, Cole, et al. - 1990
20 Molecular Biology of the Gene – Watson, Hopkins, et al. - 1987
8 Using multi-layered neural networks for learning symbolic knowledge – Sestito, Dillon - 1990
View or Download | Add to My Collection | Correct Errors