Abstract
PDF (2157 K)
E-mail Article
Add to my Quick Links
Cited By in Scopus (45)
doi:10.1016/0004-3702(95)00014-3 
Copyright © 1996 Published by Elsevier Science B.V.
Knowledge representation and inference in similarity networks and Bayesian multinets
Dan Geigera, 
, 
and David Heckermanb
Available online 9 February 1999.
Abstract
We examine two representation schemes for uncertain knowledge: the similarity network (Heckerman, 1991) and the Bayesian multinet. These schemes are extensions of the Bayesian network model in that they represent asymmetric independence assertions. We explicate the notion of relevance upon which similarity networks are based and present an efficient inference algorithm that works under the assumption that every event has a nonzero probability. Another inference algorithm is developed that works under no restriction albeit less efficiently. We show that similarity networks are not inferentially complete—namely—not every query can be answered. Nonetheless, we show that a similarity network can always answer any query of the form: “What is the posterior probability of an hypothesis given evidence?” We call this property diagnostic completeness. Finally, we describe a generalization of similarity networks that can encode more types of asymmetric conditional independence assertions than can ordinary similarity networks.
References
[1]. S. Andreassen, M. Woldbye, B. Falck and S. Andersen, MUNIN: a causal probabilistic network for interpretation of electromyographic findings. In: Proceedings IJCAI-87Milan, , Morgan Kaufmann, San Mateo, CA (1987), pp. 366–372.
[2]. J. Breese, E. Horvitz, M. Peot, R. Gay and G. Quentin, Automated decision-analytic diagnosis of thermal performance in gas turbines. In: Proceedings International Gas Turbine and Aeroengine Congress and Exposition, American Society of Mechanical Engineers, Cologne (1992).
[3]. W. Buntine, Theory refinement on Bayesian networks. In: Proceedings Seventh Conference on Uncertainty in Artificial IntelligenceLos Angeles, CA, , Morgan Kaufmann, Los Altos, CA (1991), pp. 52–60.
[4]. G.F. Cooper, Computational complexity of probabilistic inference using Bayesian belief networks. Artif. Intell. 42 (1990), pp. 393–405. Abstract | 
PDF (708 K)
| View Record in Scopus | Cited By in Scopus (297)
[5]. G.F. Cooper and E. Herskovits, A Bayesian method for constructing Bayesian belief networks from databases. In: Proceedings Seventh Conference on Uncertainty in Artificial IntelligenceLos Angeles, CA, , Morgan Kaufmann, Los Altos, CA (1991), pp. 86–94.
[6]. G.F. Cooper and E. Herskovits, A Bayesian method for the induction of probabilistic networks from data. Mach. Learn. 9 (1992), pp. 309–347. View Record in Scopus | Cited By in Scopus (608)
[7]. R. Cox, Probability, frequency and reasonable expectation. Am. J. Phys. 14 (1946), pp. 1–13.
[8]. D. Geiger, Probabilistic networks. In: S. Shapiro, Editor, Encyclopedia of Artificial Intelligence (2nd ed.),, Wiley, New York (1992), pp. 1201–1209.
[9]. D. Geiger and D. Heckerman, Separable and transitive graphoids. In: Proceedings Sixth Conference on Uncertainty in Artificial IntelligenceBoston, MA, , Association for Uncertainty in Artificial Intelligence, Mountain View, CA (1990), pp. 538–545.
[10]. D. Geiger and D. Heckerman, Learning Gaussian networks. In: Proceedings Tenth Conference on Uncertainty in Artificial IntelligenceSeattle, WA, , Morgan Kaufmann, Los Altos, CA (1994), pp. 235–243.
[11]. D. Geiger and J. Pearl, On the logic of causal models. In: Proceedings Fourth Workshop on Uncertainty in Artificial IntelligenceMinneapolis, MN, , Association for Uncertainty in Artificial Intelligence, Mountain View, CA (1988), pp. 136–147.
D. Geiger and J. Pearl also in: . In: R. Shachter, T. Levitt, L. Kanal and J. Lemmer, Editors, Uncertainty in Artificial Intelligence 4, Wiley, New York (1990), pp. 3–14.
[12]. D. Geiger, T. Verma and J. Pearl, Identifying independence in bayesian networks. Networks 20 (1990), pp. 507–534.
[13]. M. Goldbaum, B. Cote, A. Listhaus, D. Heckerman, E. Horvitz and J. Breese, IntellEye, an expert system for diagnostic ophthalmologic diseases from images of the ocular fundus. In: Investigative Ophthalmology and Visual Science (1991) Supplement .
[14]. D. Heckerman. In: Probabilistic Similarity Networks, MIT Press, Cambridge, MA (1991).
[15]. D. Heckerman, D. Geiger and D. Chickering, Learning Bayesian networks. Mach. Learn. 20 (1995), pp. 197–243. View Record in Scopus | Cited By in Scopus (449)
[16]. D. Heckerman, E. Horvitz and B. Nathwani, Toward normative expert systems, Part I: The Pathfinder project. Methods Inf. Medicine 31 (1992), pp. 90–105. View Record in Scopus | Cited By in Scopus (81)
[17]. D. Heckerman and B. Nathwani, An evaluation of the diagnostic accuracy of Pathfinder. Comput. Biomed. Res. 25 (1992), pp. 56–74. Abstract | Article | PDF (1411 K)
| View Record in Scopus | Cited By in Scopus (12)
[18]. D. Heckerman and B. Nathwani, Toward normative expert systems, Part II: Probability-based representations for efficient knowledge acquisition and inference. Methods Inf. Medicine 31 (1992), pp. 106–116. View Record in Scopus | Cited By in Scopus (27)
Heckerman and Nathwani, 1993. also in D. Heckerman and B. Nathwani. In: J. van Bemmel and A. McCray, Editors, Yearbook of Medical Informatics, International Medical Informatics Association, Rotterdam (1993), pp. 430–440.
[19]. R. Howard and J. Matheson, Influence diagrams. In: R. Howard and J. Matheson, Editors, Readings on the Principles and Applications of Decision Analysis Vol. II, Strategic Decisions Group, Menlo Park, CA (1981), pp. 721–762.
[20]. F. Jensen and S. Andersen, Approximations in Bayesian belief universes for knowledge based systems. In: Proceedings Sixth Conference on Uncertainty in Artificial IntelligenceCambridge, MA, , Morgan Kaufmann, Los Altos, CA (1990), pp. 162–169.
[21]. F. Jensen, S. Lauritzen and K. Olesen, Bayesian updating in recursive graphical models by local computations. Comput. Stat. Quart. 4 (1990), pp. 269–282.
[22]. F. Jensen, K. Olesen and S. Andersen, An algebra of Bayesian belief universes for knowledge- based systems. Networks 20 (1990), pp. 637–660.
[23]. J. Kim and J. Pearl, A computational model for causal and diagnostic reasoning in inference engines. In: Proceedings IJCAI-83 (1983), pp. 190–193. View Record in Scopus | Cited By in Scopus (44)
[24]. H. Kyburg and H. Smokler. In: Studies in Subjective Probability, Wiley, New York (1980).
[25]. W. Lam and F. Bacchus, Using causal information and local measures to learn Bayesian networks. In: Proceedings Ninth Conference on Uncertainty in Artificial IntelligenceWashington, DC, , Morgan Kaufmann, Los altos, CA (1993), pp. 243–250.
[26]. S. Lauritzen and D. Spiegelhalter, Local computations with probabilities on graphical structures and their application to expert systems. J. Roy. Stat. Soc. B 50 (1988), pp. 157–224.
[27]. B. Nathwani, D. Heckerman, E. Horvitz and T. Lincoln, Integrated expert systems and videodisc in surgical pathology: an overview. Human Path. 21 (1990), pp. 11–27. Abstract | Article | PDF (2197 K)
| View Record in Scopus | Cited By in Scopus (7)
[28]. G. Nino-Murcia and M. Shwe, An expert system for diagnosis of sleep disorders. In: M. Chase, R. Lydic and C. O'Connor, Editors, Sleep Research 20, Brain Information Service, Los Angeles, CA (1991), p. 433.
[29]. S. Olmsted, On representing and solving decision problems. In: Ph.D. Thesis, Department of Engineering-Economic Systems, Stanford University, Stanford, CA (1983).
[30]. J. Pearl, Reverend Bayes on inference engines: a distributed hierarchical approach. In: Proceedings AAAI-82Pittsburgh, PA, , AAAI Press, Menlo Park, CA (1982), pp. 133–136. View Record in Scopus | Cited By in Scopus (20)
[31]. J. Pearl, Fusion, propagation, and structuring in belief networks. Artif. Intell. 29 (1986), pp. 241–288. Abstract | PDF (2362 K)
| View Record in Scopus | Cited By in Scopus (253)
[32]. J. Pearl. In: Probabilistic Reasoning in Intelligent Systems: Networks of Plausible Inference, Morgan Kaufmann, San Mateo, CA (1988).
[33]. J. Pearl and T. Verma, A theory of inferred causation. In: J. Allen, R. Fikes and E. Sandewall, Editors, Knowledge Representation and Reasoning: Proceedings Second International Conference, Morgan Kaufmann, San Mateo, CA (1991), pp. 441–452.
[34]. L. Savage. In: The Foundations of Statistics, Dover, New York (1954).
[35]. R. Shachter, Evaluating influence diagrams. Oper. Res. 34 (1986), pp. 871–882. View Record in Scopus | Cited By in Scopus (207)
[36]. R. Shachter and C. Kenley, Gaussian influence diagrams. Manage. Sci. 35 (1989), pp. 527–550.
[37]. P. Spirtes, C. Glymour and R. Scheines. In: Causation, Prediction, and Search, Springer, New York (1993).
[38]. T. Verma and J. Pearl, Causal networks: semantics and expressiveness. In: Proceedings Fourth Workshop on Uncertainty in Artificial IntelligenceMinneapolis, MN, , Association for Uncertainty in Artificial Intelligence, Mountain View, CA (1988), pp. 352–359.
T. Verma and J. Pearl also in: . In: R. Shachter, T. Levitt, L. Kanal and J. Lemmer, Editors, Uncertainty in Artificial Intelligence 4, Wiley, New York (1990), pp. 69–76.
