Abstract
A posteriori inference is one of three kinds of inference that underlie the processing of knowledge patterns with probabilistic uncertainty in intelligent decision making systems using algebraic Bayesian networks (ABNs). In this paper, the key terms and formulations of theorems describing local a posteriori inference in algebraic Bayesian networks are given in a matrix-vector language. The main result is that matrix equations are constructed for normalizing factors appearing in the formulas of a posteriori probabilities of proposition-quanta and ideals of conjuncts. The matrix equations of local a priori inference formulated in general not only simplify the preparation of specifications of appropriate inference algorithms and make their implementation more transparent, but also open a possibility for application of classical mathematical techniques to the analysis of the properties of inference results.
Similar content being viewed by others
References
R. Bellman, Introduction to Matrix Analysis (McGraw-Hill, New York, 1960; Nauka, Moscow, 1969).
V. I. Gorodetskii, “Algebraic Bayesian networks–new paradigm of expert systems,” in Anniversary Collection of Proceedings of Institutes of the Russian Academy of Sciences Department of Informatics, Computer Science and Automation, Vol. 2 (Ross. Akad. Nauk, Moscow, 1993) [in Russian].
V. I. Gorodetskii and A. L. Tulup’ev, “Formation of noncontradicting knowledge bases with an uncertainty,” Izv. Ross. Akad. Nauk, Ser. Teor. Sist. Upr. 5, 33–42 (1997).
A. V. Sirotkin, “Computational complexity of local posteriori inference algorithms in algebraic Bayesian networks,” Tr. SPIIRAN, No. 3(18), 188–214 (2011).
A. V. Sirotkin and A. L. Tulup’ev, “Local a priori inference in algebraic Bayesian network: A set of basic algorithms,” Tr. SPIIRAN, No. 5, 100–111 (2007).
A. L. Tulup’ev, “Algebraic Bayesian networks: a local probabilistic logic inference,” in Handbook on Elements of Soft Computing (Anatoliya, St. Petersburg, 2007) [in Russian].
A. L. Tulup’ev, Algebraic Bayesian Networks: a Probabilistic Logic Approach to the Modelling of Knowledge Bases with an Uncertainty (SPIIRAN, St. Petersburg, 2000) [in Russian].
A. L. Tulup’ev, Bayesian Networks: A Probabilistic Logic Inference in Cycles (S.-Peterb. Gos. Univ., St. Petersburg, 2008) [in Russian].
A. L. Tulup’ev and A. V. Sirotkin, “Algebraic Bayesian networks: Principle of decomposition and probabilistic logic inference under uncertainty,” Inf.-Izmer. Upravl. Sist. 6 (10), 85–87 (2008).
A. L. Tulup’ev and A. V. Sirotkin, “Matrix equations for local probabilistic logic inference of truth values of elements in algebraic Bayesian networks,” Vestn. S.-Peterb. Univ., Ser. 1: Mat., Mekh., Astron., No. 3, 63–72 (2012).
A. L. Tulup’ev, S. I. Nikolenko, and A. V. Sirotkin, Bayesian Networks: A Probabilistic Logic Approach (Nauka, St. Petersburg, 2006) [in Russian].
A. L. Tulup’ev, A. V. Sirotkin, and S. I. Nikolenko, Bayesian Belief Networks: Probabilistic Logic Inference in Acyclic Directed Graphs (S.-Peterb. Gos. Univ., St. Petersburg, 2009) [in Russian].
A. A. Fil’chenkov, “Hierarchy of algebraic Bayesian network global structures as a system of graphs and hypergraphs,” Nauchno-Tekhn. Vestn. Inf. Tekhnol., Mekh. Opt. 13 (1), 75–81 (2013).
N. J. Nilsson, “Probabilistic Logic,” Artif. Intell. 47, 71–87 (1986).
Author information
Authors and Affiliations
Corresponding author
Additional information
Original Russian Text © A.L. Tulupyev, A.V. Sirotkin, A.A. Zolotin, 2015, published in Vestnik Sankt-Peterburgskogo Universiteta. Seriya 1. Matematika, Mekhanika, Astronomiya, 2015, No. 3, pp. 391–398.
About this article
Cite this article
Tulupyev, A.L., Sirotkin, A.V. & Zolotin, A.A. Matrix equations for normalizing factors in local a posteriori inference of truth estimates in algebraic Bayesian networks. Vestnik St.Petersb. Univ.Math. 48, 168–174 (2015). https://doi.org/10.3103/S1063454115030073
Received:
Published:
Issue Date:
DOI: https://doi.org/10.3103/S1063454115030073