Rule base management using meta knowledge

ABSTRACT

This paper describes the rule base management strategy of an expert system environment. The environment includes a set of integrated tools which facilitate acquisition, manipulation and maintenance of knowledge. The rule base management component of the system, called RBM, assists these tasks by organizing global semantic information within the rule base. RBM extracts this semantic information from the texts included in “rule structures” and builds a semantic network of the concepts found in the rule base. The rule base is then divided into rulesets which are clusters of rules that refer to the same atomic concept. Construction of this meta knowledge is achieved through a keyword matching mechanism. The paper includes a brief description of the RBM system, the dictionary it uses for building meta-level knowledge, and its keyword matching technique.

REFERENCES

• 1: Clark, K L and McCabe, F G, "Prolog a language for Implementing expert systems," m Machine Intelhgence 10, eds D Mlchm and Y H Pao, 1982
• 2: Forgy, C L, "OPS4 User's Manual," Technical Report CMU-CS-79-132, Carnegle-Mellon University, Department of Computer Scmnce, 1979
• 3: Bobrow, D G, Stefik, M The Loops Manual, XEROX PARC, 1983
• 4: Harandl, M T, "Knowledge-based program debugging A heuristic model", Proceedings of IEEE Conference on Software Development Tools, Techniques, and Alternatives (SOFTFAIR-I), Arhngton, VA (1983)
• 5: M T Harandi, The architecture of an expert system environment, 5th international workshop, Vol. 1 on Expert systems & their applications, p.555-572, December 1986, Avignon, France
• 6: Harandl, M T, "A tree-based knowledge representatmn scheme for diagnostic expert systems," Proceedings of the 2nd Conference on Intelhgent Systems and Machines, Rochester, MI, (1084)
• 7: Davis, R #lnteraehve Transfer of Expertise Acqulsttlon of New Inference Rules", Arhficlal Intelligence vol 12, pp 121-157, 1979
• 8: Randall Davis , Douglas B. Lenat, Knowledge-Based Systems in Artificial Intelligence: 2 Case Studies, McGraw-Hill, Inc., New York, NY, 1982
• 9: Shorthffe, E H, "Computer-Based Medical Consultations MYCIN', American Elsevier Pubhshlng Co, New York, 1976
• 10: A}kens, J S, "Prototypes and Production Rules A Knowledge Representatmn for Computer Consultatmns ", Stanford Heuriutlc Programming Project, Report No STAN-CS-80-814, (1980)
• 11: Tenczar, P, and Golden, W, "Spelling, Word, And Concept Recognltion", CERL Report X-35, Computer-based Educatmn Research Laboratory, University of Ilhnom, Urbana, Ilhnom, October 1972
• 12: Roberts, C, "Partial-Match Retrieval via the Method of Superimposed Codes", Proceedsnga of the IEEE, Vol 67, No 12, December 1979

INDEX TERMS

Classification:
  H. Information Systems
  H.2 DATABASE MANAGEMENT

  I. Computing Methodologies
  I.2 ARTIFICIAL INTELLIGENCE
      I.2.1 Applications and Expert Systems
          Subjects: Natural language interfaces
      I.2.4 Knowledge Representation Formalisms and Methods
          Subjects: Representations (procedural and rule-based); Semantic networks

General Terms:
Design, Languages, Management, Performance