Abstract
Directed evidential networks (DEVNs) can be seen, at present, as an extremely powerful graphical tool for representing and reasoning with uncertain knowledge in the framework of evidence theory.
The main purpose of this paper is twofold. Firstly, it introduces hybrid directed evidential networks which generalize the standard DEVNs. Secondly, it presents an algorithm for performing inference over singly-connected hybrid evidential networks.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
Notes
- 1.
In Bayesian networks, conditional probabilities are specified per child node, i.e. for each node given all its parents. However, in ENCs, if there are two edges going from nodes A and B to their common child node C, then two conditional belief function distributions have to be defined: a distribution for C given A and another one for C conditionally to B.
- 2.
The notations m[N\(_\mathrm{{k}}\)](N\(_\mathrm{{i}}\)) and m\(^{{\small \mathrm{{N}}_\mathrm{{i}}}}[{\small \mathrm{{N}}_\mathrm{{k}}}]\) used throughout this paper correspond to the classical notation m(N\(_\mathrm{{i}}\) \(\mid \)N\(_\mathrm{{k}}\)).
- 3.
Case 1 occurs when the child node \({\mathrm{{N}}_\mathrm{{i}}}\) is associated with conditionals specified per single parent.
- 4.
Case 2 occurs when the child node \({\mathrm{{N}}_\mathrm{{i}}}\) has one conditional distribution defined for all its parents.
- 5.
Case 3 occurs when \({\mathrm{{N}}_\mathrm{{j}}}\) has one or more child nodes associated with conditionals specified per single parent.
- 6.
Case 4 occurs when \({\mathrm{{N}}_\mathrm{{j}}}\) has one or more child nodes associated with conditionals specified for all the parents.
References
Ben Yaghlane, B., Mellouli, K.: Inference in directed evidential networks based on the transferable belief model. IJAR 48(2), 399–418 (2008)
Ban Yaghlane, B., Mellouli, K.: Updating directed belief networks. In: Hunter, A., Parsons, S. (eds.) ECSQARU 1999. LNCS (LNAI), vol. 1638, pp. 43–54. Springer, Heidelberg (1999)
Dempster, A.P.: Upper and lower probabilities induced by a multivalued mapping. Ann. Math. Stat. 38, 325–339 (1967)
Laâmari, W., Ben Yaghlane, B.: Reasoning in singly-connected directed evidential networks with conditional beliefs. In: Likas, A., Blekas, K., Kalles, D. (eds.) SETN 2014. LNCS, vol. 8445, pp. 221–236. Springer, Heidelberg (2014)
Pearl, J.: Probabilistic Reasoning in Intelligent Systems: Networks of Plausible Inference. Morgan Kaufmann, San Mateo (1988)
Shafer, G.: A Mathematical Theory of Evidence. Princeton University Press, Princeton (1976)
Shenoy, P.P.: Valuation networks and conditional independence. In: Uncertainty in Artificial Intelligence, pp. 191–199 (1993)
Smets, Ph.: Belief function: the disjunctive rule of combination and the generalized Bayesian theorem. Int. J. Approx. Reasoning 9, 1–35 (1993)
Smets, Ph., Kennes, R.: The transferable belief model. Artif. Intell. 66, 191–234 (1994)
Xu, H., Smets, Ph.: Evidential reasoning with conditional belief functions. In: Heckerman, D., et al. (eds.) Proceedings of Uncertainty in Artificial Intelligence (UAI 1994), pp. 598–606. Morgan Kaufmann, San Mateo (1994)
Zadeh, L.A.: Fuzzy sets. Inf. Control 8, 338–353 (1965)
Author information
Authors and Affiliations
Corresponding author
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2015 Springer International Publishing Switzerland
About this paper
Cite this paper
Laâmari, W., Ben Yaghlane, B. (2015). Propagation of Belief Functions in Singly-Connected Hybrid Directed Evidential Networks. In: Beierle, C., Dekhtyar, A. (eds) Scalable Uncertainty Management. SUM 2015. Lecture Notes in Computer Science(), vol 9310. Springer, Cham. https://doi.org/10.1007/978-3-319-23540-0_16
Download citation
DOI: https://doi.org/10.1007/978-3-319-23540-0_16
Published:
Publisher Name: Springer, Cham
Print ISBN: 978-3-319-23539-4
Online ISBN: 978-3-319-23540-0
eBook Packages: Computer ScienceComputer Science (R0)