Diagnostic problem-solving with causal chaining. (English) Zbl 0635.68113
A part of the inference methods known in the field of diagnostic problem- solving models human processes of reasoning. The methods named by the authors “abductive”, i.e. inferring the best or most plausible explanations for a given set of facts, attempt to do expert-level performance, in a fashion that mimics the human diagnostic problem- solving.
This paper extends the authors’ previous research in a formal model of abductive diagnostic problem-solving named by them “parsimonious covering theory”. The extensions generalize the representation of diagnostic knowledge by adding intermediate states in the underlying causal network allowing causal chaining. Thus this theory can handle a much broader range of real-world diagnostic problems.
After a short introduction, the diagnostic problem and its solution is formally defined in the second section. To simplify presentation of the algorithms in subsequent sections, an algebra over generator sets (which are representations of the solutions to diagnostic problems) is presented in the next section. In the fourth section an algorithm is developed to solve the simplest case of diagnostic problems, named bipartite. More general algorithms for abductive problem solving involving causal chaining are successively presented in the next two sections. Further research directions and applications currently under development are the subjects of the last section.
The paper is rigorously written, proofs of all the results are given in the Appendix. We consider that this paper fulfils the authors’ original intention to develop a theoretical foundation for association-based abductive inference methods.
This paper extends the authors’ previous research in a formal model of abductive diagnostic problem-solving named by them “parsimonious covering theory”. The extensions generalize the representation of diagnostic knowledge by adding intermediate states in the underlying causal network allowing causal chaining. Thus this theory can handle a much broader range of real-world diagnostic problems.
After a short introduction, the diagnostic problem and its solution is formally defined in the second section. To simplify presentation of the algorithms in subsequent sections, an algebra over generator sets (which are representations of the solutions to diagnostic problems) is presented in the next section. In the fourth section an algorithm is developed to solve the simplest case of diagnostic problems, named bipartite. More general algorithms for abductive problem solving involving causal chaining are successively presented in the next two sections. Further research directions and applications currently under development are the subjects of the last section.
The paper is rigorously written, proofs of all the results are given in the Appendix. We consider that this paper fulfils the authors’ original intention to develop a theoretical foundation for association-based abductive inference methods.
Reviewer: G.Curelet-Balan
MSC:
| 68T20 | Problem solving in the context of artificial intelligence (heuristics, search strategies, etc.) |
Keywords:
diagnostic expert systems; networks of causal associations; problem- solving; parsimonious covering theory; abductive inferenceReferences:
| [1] | and , ”An overview of methods for computer-assisted medical decision making”, in Computer-Assisted Medical Decision Making, and (Eds.), Vol. 1, Springer-Verlag, NY, 1985, pp. 3–45. · doi:10.1007/978-1-4613-8554-7_1 |
| [2] | ”On the mechanization of abductive logic”, Proc. of International Joint Conference on Artificial Intelligence, IJCAI, 1973. pp. 147–152. |
| [3] | ”Heuristic methods for imposing structure on ill-structured problems: the structuring of medical diagnostics”, Artificial Intelligence in Medicine, P. Szolovits, (Ed.), 1982, pp. 119–190. |
| [4] | Miller, New England J. of Medicine 307 pp 468– (1982) |
| [5] | Pauker, Am. J. Med. 60 pp 981– (1976) |
| [6] | Knowledge-Based Decision Support Systems: Development through KMS, Tech, Report TR-1121 Dept. of Comp. Sci., Univ. of Maryland, Oct., 1981. |
| [7] | , and , ”Diagnostic expert systems based on a set covering model”, Int. J. Man-Machine Studies, Nov. 1983, pp. 437–460. |
| [8] | , and , ”Assembling the best explanation”, IEEE Workshop on Principles of Knowledge-Based Systems, Denver, CO. Dec., 1984. |
| [9] | and , ”ARROWSMITH-P: A prototype expert system for software engineering management”. Proc. Expert Systems in Government Symposium, (Ed.), Mclean, Virginia, 1985 |
| [10] | and , Introduction to Artificial Intelligence, Addision-Wesley, Reading, MA, 1985, Chapters 8, 10. · Zbl 0605.68083 |
| [11] | Abduction and Induction, Dover, 1955. |
| [12] | ”Abductive inference”, Expert Systems in Government Symposium, Oct. 1985, pp. 484–489. |
| [13] | Ben-Bassat, IEEE Transactions on Pattern Analysis and Machine Intelligence 2 pp 148– (1980) |
| [14] | Reggia, Information Sciences 37 pp 227– (1985) |
| [15] | A Formalization of Parsimonious Covering and Probabilistic Reasoning in Abductive Diagnostic Inference, Tech. Report TR-1615 (Ph.D. Dissertation), Dept. of Comp. Sci., Univ. of Maryland, Jan. 1986. |
| [16] | and , ”A probabilistic causal model for diagnostic problemsolving”, IEEE Transactions on Systems, Man and Cybernetics, in press, 1987. |
| [17] | ”A theory of diagnosis from first principles”, TR-187/86, Dept. of Computer Science, University of Toronto, Dec. 1985. |
| [18] | and , ”Reasoning about multiple faults”, submitted, 1986. |
| [19] | NETL-A System for Representing and Using Real-World Knowledge, MIT Press, Cambridge, Massachusetts, 1979. · Zbl 0444.68083 |
| [20] | Thagard, Journal of Philosophy 75 pp 76– (1978) |
| [21] | and , ”Plausibility of diagnostic hypotheses: the nature of simplicity”, Proc. of the National Conference on Artificial Intelligence, 1986, pp. 140–145. |
| [22] | and , ”Relationship between deductive and abductive inference in knowledge-based diagnostic problem solving”, Proc. First Intl. Workshop on Expert Database Systems, (Ed.), Kiawah Island, Sc., Oct, 1984, pp. 500–509. |
| [23] | , , and , ”Answer justification in diagnostic expert systems”, IEEE Trans. on Biomed. Eng., BME-32, April 1985, pp. 263–272. |
| [24] | Yager, Int. Journal of Man-Machine Studies 23 pp 539– (1985) |
| [25] | , , , , , and , ”Plausible reasoning during neurological problem-solving”, Proc. Fifth World Congress on Medical Informatics, Washington, DC, Oct. 1986, pp. 17–21. |
| [26] | and , ”Abduction in discourse processing: A parsimonious covering model”, Proc. Second Symp. on the Role of Language in Problem-Solving, (1986), in press. |
This reference list is based on information provided by the publisher or from digital mathematics libraries. Its items are heuristically matched to zbMATH identifiers and may contain data conversion errors. In some cases that data have been complemented/enhanced by data from zbMATH Open. This attempts to reflect the references listed in the original paper as accurately as possible without claiming completeness or a perfect matching.
