Goodman, Rodney M. and Higgins, Charles M. and Miller, John W. and Smyth, Padhraic (1992) Rule-based neural networks for classification and probability estimation. Neural Computation, 4 (6). pp. 781-804. ISSN 0899-7667. http://resolver.caltech.edu/CaltechAUTHORS:GOOnc92
- Published Version
See Usage Policy.
Use this Persistent URL to link to this item: http://resolver.caltech.edu/CaltechAUTHORS:GOOnc92
In this paper we propose a network architecture that combines a rule-based approach with that of the neural network paradigm. Our primary motivation for this is to ensure that the knowledge embodied in the network is explicitly encoded in the form of understandable rules. This enables the network's decision to be understood, and provides an audit trail of how that decision was arrived at. We utilize an information theoretic approach to learning a model of the domain knowledge from examples. This model takes the form of a set of probabilistic conjunctive rules between discrete input evidence variables and output class variables. These rules are then mapped onto the weights and nodes of a feedforward neural network resulting in a directly specified architecture. The network acts as parallel Bayesian classifier, but more importantly, can also output posterior probability estimates of the class variables. Empirical tests on a number of data sets show that the rule-based classifier performs comparably with standard neural network classifiers, while possessing unique advantages in terms of knowledge representation and probability estimation.
|Additional Information:||© 1992 Massachusetts Institute of Technology. Received 19 April 1991; accepted 26 March 1992. Posted Online March 13, 2008. This work is supported in part by Pacific Bell, in part by the Army Research Office under Contract DAAL03-89-K-0126, and in part by DARPA under contract AFOSR-90-0199. Part of this research was carried out by the Jet Propulsion Laboratory, California Institute of Technology, under a contract with the National Aeronautics and Space Administration. The authors thank David Aha of U.C. Irvine for providing the voting data set, and Olvi Mangasarian of the University of Wisconsin for providing the medical diagnosis data.|
|Usage Policy:||No commercial reproduction, distribution, display or performance rights in this work are provided.|
|Deposited By:||Tony Diaz|
|Deposited On:||25 Jun 2009 16:36|
|Last Modified:||26 Dec 2012 10:53|
Repository Staff Only: item control page