Title :
Reasoning with belief structures: non monotonic compatibility relations
Author :
Yager, Ronald R.
Author_Institution :
Iona Coll., New Rochelle, NY, USA
Abstract :
Our concern is with getting information about one variable using knowledge of a second variable. We use of Dempster-Shafer belief structure to provide a representation, in a granular form, of knowledge about a variable when we have some uncertainty in our knowledge. We introduce compatibility relations for representing the possible joint values of two variables. Our inference scheme involves using a compatibility relation and a belief structure on one variable to infer a belief structure on the second variable. Our concern is with the representation of different types of compatibility relations. A distinction introduced is between monotonic and non-monotonic compatibility relations. We provide for the representation of both types.
Keywords :
belief networks; nonmonotonic reasoning; statistical distributions; uncertainty handling; Dempster-Shafer belief structure; inference scheme; knowledge granular form; monotonic reasoning; nonmonotonic compatibility relation; Educational institutions; Expert systems; Hafnium; Input variables; Probability distribution; Random variables; Uncertainty;
Conference_Titel :
Granular Computing, 2005 IEEE International Conference on
Print_ISBN :
0-7803-9017-2
DOI :
10.1109/GRC.2005.1547238