Zusammenfassung
Die Repräsentation und Verarbeitung unsicheren Wissens in einer mathematisch korrekten und berechenbaren Weise ist eines der zentralen Anliegen der Künstlichen Intelligenz. Am Beispiel der schon berühmt gewordenen Studentin Léa Sombé aus [27] zeigen wir in dieser Arbeit, wie subjektives, unvollständiges Wissen in einer probabilistischen Wissensbasis dargestellt und als Grundlage für Inferenzen dienen kann. Dies wird durch entropie-optimalen Abgleich von Verteilungen erreicht. Diese in der Expertensystem-Shell SPIRIT implementierte Methodik zeichnet sich nicht nur durch gute Berechenbarkeit aus, sondern gestattet auch dank der hier aufgezeigten Kompatibilität mit der Konditionallogik eine logisch profunde Informationsverarbeitung.
Abstract
In this paper, the famous Léa Sombé-example of [27] is re-examined by methods combining cross-entropy minimization with probabilistic conditional logic. Thus a knowledge base is built up which allows easy computations and inferences in a logically sound way. The concept is realized by the probabilistic expert system shell SPIRIT which is presented here, too. So the aim of this paper affects as much practical aspects as it concerns logical foundations of knowledge representation. As the Sombé-example illustrates, even incomplete knowledge based on subjective probabilities or statistical data may be represented and dealt with adequately.
Article PDF
Avoid common mistakes on your manuscript.
Literatur
Adams EW (1975) The Logic of Conditionals. D. Reidel, Dordrecht
Calabrese PG (1991) Deduction and interference using conditional logic and probability. In: Goodman IR et al. (eds) Conditional Logic in Expert Systems. North-Holland, Amsterdam, pp 71–100
Csiszár I (1975) I-divergence Geometry of Probability Distributions and Minimization Problems. Ann Probab 3: 146–158
De Finetti B (1937) La prévision, ses lois logiques et ses sources subjectives. Ann Inst H Poincaré 7 (1937) translated by Kyburg H Jr. In: Kyburg H Jr., Smokler H (eds) (1964) Studies in Subjective Probability. Wiley, New York, pp 95–158
Dubois D, Prade H (1991) Conditioning, Non-monotonic Logic and Non-standard Uncertainty Models. In: Goodman IR et al. (eds), Conditional Logic in Expert Systems. North-Holland, Amsterdam, pp 115–158
Gerla G (1994) The Probability that Tweety is Able to Fly. Int J Intelligent Systems 9: 403–409
Goodman IR, Nguyen HT (1988) Conditional objects and the modeling of uncertainties. In: Gupta MM, Yamakawa T (eds) Fuzzy Computing. North-Holland, Amsterdam, pp 119–138
Jaynes ET (1983) Papers on Probability, Statistics and Statistical Physics. Rosenkrantz RD (ed) D. Reidel, Dordrecht Holland
Kullback S (1968) Information Theory and Statistics. Dover, New York
Lauritzen SL (1982) Lectures on Contingency Tables. Aalborg University Press, Denmark
Lauritzen SL, Spiegelhalter DJ (1988) Local Computations with Probabilities in Graphical Structures and Their Applications to Expert Systems. J R Statist Soc B 50(2): 415–448
Lemmer JF (1983) Generalized Bayesian updating of incompletely specified distributions. Large Scale Syst 5: 51–68
Lemmer JF, Barth SW (1982) Efficient minimum information updating for Bayesian inferencing in expert systems. In: Proceedings of the National Conference on Artificial Intelligence. Morgan Kauffmann, Pittsburgh, PA, Los Altos, CA, pp 424–427
Makinson D (1989) General Theory of Cumulative Inference. In: Reinfrank M (ed) Non-monotonic Reasoning. Springer, Berlin, pp 1–18
Malvestuto FM (1986) Decomposing complex contingency tables to reduce storage requirements. In: Proc. 3rd Intern. Workshop on Scientific and Statistical Database Management, pp 66–71
Meyer CH, Kern-Isberner G, Rödder W (1996) Analyse medizinisch-soziologischer Daten mittels eines probabilistischen Expertensystems. In: Kleinschmid P et al. (Hrsg) Operations Research Proceedings 1995. Springer, Berlin et al., pp 347–352
Nilsson NJ (1986) Probabilistic Logic. Artificial Intelligence 28(1): 71–87
Pearl J (1988) Probabilistic Reasoning in Intelligent Systems. Morgan Kauffmann, San Mateo, CA
Reidmacher HP (1994) Logisches Schließen bei Unsicherheit. Europäische Hochschulschriften, Reihe 5, Bd. 1519, Frankfurt am Main
Reidmacher HP, Kern-Isberner G (1996) Interpreting a Contingency Table by Rules. Int J Intelligent Syst 11: 327–346
Rödder W (1994) Symmetrical Probabilistic Intensional Reasoning in Inference Networks in Transition. In: Werners B, Gabriel R (Hrsg) Operations Research. Springer, Berlin Heidelberg New York, pp 129–145
Rödder W, Longgui X (1993) SPIRIT — Die Behandlung logischer Funktionen in einer probabilistischen Wissensbasis. In: Hansmann K-W et al. (Hrsg) Operations Research Proceedings 1992. Springer, Berlin et al., pp 462–469
Shannon CE (1948) A mathematical theory of communication. Bell Syst Tech J 27: 379–423; 623–656.
Shore JE (1986) Relative Entropy, Probabilistic Inference and AI. In: Kanal et al. (eds) Uncertainty in AI. North-Holland, Amsterdam, pp 211–215
Shore JE, Johnson RW (1980) Axiomatic Derivation of the Principle of Maximum Entropy and the Principle of Minimum Cross-Entropy. IEEE Transact Inf Theory IT-26(1): 26–37
Shortliffe E (1976) Computer based medical consultation: MYCIN. American Elsevier Publishing Co, Amsterdam
Sombé L (1992) Schließen bei unsicherem Wissen in der Künstlichen Intelligenz. Vieweg, Braunschweig, Wiesbaden
Spiegelhalter DJ, Dawid AP, Lauritzen SL, Cowell RG (1993) Bayesian analysis in expert systems. Statist Sci 8: 219–247
Zimmermann H-J (1990) Fuzzy Set Theory — and its Applications (Rev. Ed.). Boston, Dordrecht, Lancaster
Author information
Authors and Affiliations
Rights and permissions
About this article
Cite this article
Rödder, W., Kern-Isberner, G. Léa Sombé und entropie-optimale Informationsverarbeitung mit der Expertensystem-Shell SPIRIT. OR Spektrum 19, 41–46 (1997). https://doi.org/10.1007/BF01539807
Received:
Accepted:
Published:
Issue Date:
DOI: https://doi.org/10.1007/BF01539807