Abstract
This paper compares the expressions obtained from an analysis of a problem involving conflicting evidence when using Dempster's rule of combination and conditional probabilities. Several results are obtained showing if and when the two methodologies produce the same results. The role played by the normalizing constant is shown to be tied to prior probability of the hypothesis if equality is to occur. This forces further relationships between the conditional probabilities and the prior. Ways of incorporating prior information into the Belief function framework are explored and the results are analyzed. Finally a new method for combining conflicting evidence in a belief function framework is proposed. This method produces results more closely resembling the probabilistic ones.
This work is supported by NSERC operating grant #A4515.
This is a preview of subscription content, log in via an institution to check access.
Preview
Unable to display preview. Download preview PDF.
References
Charniak, E. (1983) “The Bayesian Basis of Common Sense Medical Diagnosis”. Proc. of the National Conference on Artificial Intelligence, Wash., D.C., pp. 70–73.
Chatolic, P., Dubois, D., Prade, H. (1987). “An Approach to Approximate Reasoning Based on Dempster Rule of Combination”, International Journal of Expert Systems, pp. 67–85.
Dempster, A.P. (1968). “A Generalization of Bayesian Inference”, Journal of the Royal Statistical Society, Series B, 30, pp. 205–247.
Dubois, D., Prade, H. (1988). “Representation and Combination of Uncertainty with Belief Functions and Possibility Measurements”, Computational Intelligence, Vol. 4, pp. 244–264.
Dubois, D., Prade, H. (1985). “Combination and Propagation of Uncertainty with Belief Functions”, IJCAI, 111–113.
Fagin, R., Halpern, J. (1989). “Uncertainty, Belief, and Probability”, IJCAI'89, Proceedings, pp. 1161–1167.
Lindley, D.V. (1985). “Making Decisions”, 2nd Edition, John Wiley.
McLeish, M. (March, 1989). “Extreme Probabilities and Non-Monotonic Reasoning in Probabilistic Logic, Dempster-Shafer and Certainty Factor Models”, submitted, Journal of Approximate Reasoning.
McLeish, M. (1989). “Further Work on Non-monotonicity in the Framework of Probabilistic Logic with an Extension to Dempster-Shafer Theory”, Machine Intelligence and Pattern Recognition, Vol. 8, pp. 23–34.
McLeish, M. (1990). “A Model for Non-Monotonic Reasoning Using Dempster's Rule”, Proceedings of the 6th Uncertainty Management Conference, Cambridge, Mass., pp. 518–528.
Pearl, J. (1988). “Probabilistic Reasoning in Intelligent Systems”, Morgan Kaufmann.
Pearl, J. (1990). “Reasoning with Belief Functions; an Analysis of Compatibility”, Journal of Approximate Reasoning, Vol. 4, pp. 363–389.
Shafer, G. (1979). “A Mathematical Theory of Evidence”, Princeton University Press, Princeton, NJ.
Zadeh, L.A. (1986). “A Simple View of the Dempster Shafer Theory of Evidence and its Implication for the Rule of Combination”, A.I. Magazine, 7, pp. 85–90.
Author information
Authors and Affiliations
Editor information
Rights and permissions
Copyright information
© 1991 Springer-Verlag Berlin Heidelberg
About this paper
Cite this paper
Deutsch-McLeish, M. (1991). A study of probabilities and belief functions under conflicting evidence: Comparisons and new methods. In: Bouchon-Meunier, B., Yager, R.R., Zadeh, L.A. (eds) Uncertainty in Knowledge Bases. IPMU 1990. Lecture Notes in Computer Science, vol 521. Springer, Berlin, Heidelberg. https://doi.org/10.1007/BFb0028135
Download citation
DOI: https://doi.org/10.1007/BFb0028135
Published:
Publisher Name: Springer, Berlin, Heidelberg
Print ISBN: 978-3-540-54346-6
Online ISBN: 978-3-540-47580-4
eBook Packages: Springer Book Archive