Title :
Propagating multi-observations in directed belief networks
Author :
Khalfallah, Fédia ; Yaghlane, Boutheina Ben ; Mellouli, Khaled
Author_Institution :
LAROID, Inst. Superieur de Gestion de Tunis, Tunisia
Abstract :
Deals with the task of propagating new evidence through belief networks. An important feature of these graphical networks is that they implicitly define conditional independence relationships among the variables. The proposed algorithm permits us to directly use conditional belief functions in evidential reasoning systems and allows the propagation of several observations occurring at the same time
Keywords :
belief networks; case-based reasoning; conditional belief functions; conditional independence relationships; directed belief networks; evidence propagation; evidential reasoning systems; multi-observations propagation; Bayesian methods; Computer networks; Intelligent networks; Uncertainty;
Conference_Titel :
Systems, Man, and Cybernetics, 1998. 1998 IEEE International Conference on
Conference_Location :
San Diego, CA
Print_ISBN :
0-7803-4778-1
DOI :
10.1109/ICSMC.1998.725028
