Combining Evidence in the Extended Dempster-Shafer Theory
The Dempster-Shafer (D-S) theory of evidence generalizes Bayesian probability theory, by providing a coherent representation for ignorance (lack of evidence). However, uncertain relationships between evidence and hypotheses bearing on this evidence are difficult to represent in applications of the theory. Yen  extended the theory by introducing a probabilistic mapping that uses conditional probabilities to express these uncertain relationships, and developed a three-step method for combining evidence from different evidential sources. We present a simpler, one-step method for combining evidence based on the extended D-S theory, and prove that it gives the same results.
Unable to display preview. Download preview PDF.
- Yen, J., “A reasoning model based on an extended Dempster-Shafer theory”, Proceedings aaai-86 125–131.Google Scholar
- Duda, R. O., P. E. Hart, and N. J. Nilsson,“Subjective Bayesian methods for rule-based inference systems”, Proceedings 1976 National Computer Conference, AFIPS, 45 (1976) 1075–1082.Google Scholar
- Grosof, B. N.,“Evidential confirmation as transformed probability”, Proceedings of the AAAI/IEEE Workshop on Uncertainty and Probability in Artificial Intelligence, 1985, pp. 185–192.Google Scholar
- Heckerman, D., “A probabilistic interpretation for MYCIN’s certainty factors”, Proceedings of the AAAI/IEEE Workshop on Uncertainty and Probability in Artificial Intelligence, 1985, pp. 9–20.Google Scholar
- Loui, R., J. Feldman, H. Kyburg,“Interval-based decisions for reasoning systems”, Proceedings of the AAAI/IEEE Workshop on Uncertainty and Probability in Artificial Intelligence, 1985, pp. 193–200.Google Scholar
- Hudlicka E., Lesser V. R., Pavlin J. Rewari A, “Design of a Distributed Fault Diagnosis System”, Technical Report 86–63, COINS Department, University of Massachusetts, Amherst, Massachusetts.Google Scholar
- Shafer, G., “A Mathematical Theory of Evidence”, Princeton University Press, Princeton, New Jersey, 1976.Google Scholar
- Yen, J., “GERTIS: A Dempster-Shafer approach to diagnosing hierarchical hypotheses”, Communications of the ACM, 1989 573–585.Google Scholar