Title :
Study of interval-valued belief combination with conflicting evidence
Author :
Chang, L.W. ; Kashyap, R.L.
Author_Institution :
US Naval Res. Lab., Washington, DC, USA
Abstract :
A novel mathematical procedure for combining conflicting evidence which is represented in the interval form is presented. A methodology to construct combination rules which obey a set of essential properties is proposed. The method is based on a geometric model. The authors compare results obtained from Dempster´s rule, the interval Bayes rule, and the proposed combination rule with both conflicting and nonconflicting data, and show that the values generated by the proposed combining rule are in tune with the authors´ intuition
Keywords :
Bayes methods; computational geometry; inference mechanisms; combination rules; conflicting evidence; geometric model; interval Bayes rule; interval-valued belief combination; nonconflicting data; novel mathematical procedure; Bayesian methods; Laboratories; Measurement uncertainty; Possibility theory; Solid modeling;
Conference_Titel :
Tools for Artificial Intelligence, 1990.,Proceedings of the 2nd International IEEE Conference on
Conference_Location :
Herndon, VA
Print_ISBN :
0-8186-2084-6
DOI :
10.1109/TAI.1990.130428