Refinement of Uncertain Rule Bases via Reduction

Refining deep (multilayer) rule bases of an expert system with uncertainty to cover a set of new examples can be very difficult (NP-hard). We analyze refinement via reduction, an approach first proposed by Ginsberg, who claimed that this approach eases the complexity of refining rule bases without uncertainty. We outline a model of rule bases with uncertainty, and give necessary and sufficient conditions on uncertainty combination functions that permit reduction from deep to flat (nonchaining) rule bases. We prove that reduction cannot be performed with most commonly used uncertainty combination functions. However, we show that there is a class of reducible rule bases in which the strength refinement problem is NP-hard in the deep rule base, reduction is polynomial, and the flat rule base can be refined in polynomial time. This result also allows polynomial refinement of practical expert systems in the form of rule deletion. Thus, our results provide some theoretical evidence that refinement via reduction is feasible.

[1]  Richard E. Korf,et al.  Toward a Model of Representation Changes , 1980, Artif. Intell..

[2]  E H Shorthffe,et al.  Computer-based medical consultations mycin , 1976 .

[3]  David S. Johnson,et al.  Computers and Intractability: A Guide to the Theory of NP-Completeness , 1978 .

[4]  Allen Ginsberg.  Knowledge-Base Reduction: A New Approach to Checking knowledge Bases for Inconsistency and Redundancy , 1988, AAAI.

[5]  John R. Anderson,et al.  MACHINE LEARNING An Artificial Intelligence Approach , 2009 .

[6]  David C. Wilkins,et al.  Improving the Performance of Inconsistent Knowledge Bases via Combined Optimization Method , 1991, ML.

[7]  Allen Ginsberg,et al.  Knowledge Base Refinement and Theory Revision , 1989, ML.

[8]  Stan Matwin,et al.  Constructive Inductive Logic Programming , 1993, IJCAI.

[9]  Tom Michael Mitchell,et al.  Explanation-based generalization: A unifying view , 1986 .

[10]  Ryszard S. Michalski,et al.  A theory and methodology of inductive learning , 1993 .

[11]  D. Benjamin.  Change of Representation and Inductive Bias , 1989 .

[12]  Donald W. Loveland,et al.  On the complexity of belief network synthesis and refinement , 1992, Int. J. Approx. Reason..

[13]  Marco Valtorta.  Some results on the computational complexity of refining confidence factors , 1991, Int. J. Approx. Reason..

[14]  J. Ross Quinlan,et al.  Inferno: A Cautious Approach To Uncertain Inference , 1986, Comput. J..

[15]  Bruce G. Buchanan,et al.  The MYCIN Experiments of the Stanford Heuristic Programming Project , 1985 .

[16]  Mark A. Musen,et al.  Automated Support for Building and Extending Expert Models , 2005, Machine Learning.

[17]  Marco Valtorta.  More Results on the Complexity of Knowledge Base Refinement: Belief Networks , 1990, ML.

[18]  Saul Amarel.  Expert behaviour and problem representations , 1984 .

[19]  Shijie Wang,et al.  On the conversion of rule bases into belief networks , 1992, SAC '92.

[20]  Thomas G. Dietterich,et al.  A Comparative Review of Selected Methods for Learning from Examples , 1983 .

[21]  Judea Pearl,et al.  Probabilistic reasoning in intelligent systems - networks of plausible inference , 1991, Morgan Kaufmann series in representation and reasoning.

[22]  Michael Kearns,et al.  Computational complexity of machine learning , 1990, ACM distinguished dissertations.

[23]  Ian Benson.  Prospector: an expert system for mineral exploration , 1986 .

[24]  Allen Ginsberg,et al.  Automatic Refinement of Expert System Knowledge Bases , 1988 .

[25]  Masaki Togai,et al.  Expert System on a Chip: An Engine for Real-Time Approximate Reasoning , 1986, IEEE Expert.

[26]  A Elithorn,et al.  ARTIFICIAL AND HUMAN INTELLIGENCE , 1984 .

[27]  Edward H. Shortliffe,et al.  Rule Based Expert Systems: The Mycin Experiments of the Stanford Heuristic Programming Project (The Addison-Wesley series in artificial intelligence) , 1984 .

[28]  Carl H. Smith,et al.  Inductive Inference: Theory and Methods , 1983, CSUR.

[29]  RICHARD 0. DUDA,et al.  Subjective bayesian methods for rule-based inference systems , 1899, AFIPS '76.

[30]  Charles X. Ling,et al.  Inductive Learning from Good Examples , 1991, IJCAI.

[31]  R. Bareiss,et al.  Supporting Start-to-Finish Development of Knowledge Bases , 2005, Machine Learning.

[32]  Richard O. Duda,et al.  Subjective bayesian methods for rule-based inference systems , 1976, AFIPS '76.

[33]  Marin Marinov,et al.  A Symbolic Model for Learning the Past-Tenses of English Verbs , 1993, IJCAI.