Advertisement
Published in:
2011 | OriginalPaper | Chapter
Experimental Comparative Study of Compilation-Based Inference in Bayesian and Possibilitic Networks
Authors : Raouia Ayachi, Nahla Ben Amor, Salem Benferhat
Published in: Fuzzy Logic and Applications
Publisher: Springer Berlin Heidelberg
Activate our intelligent search to find suitable subject content or patents.
Select sections of text to find matching patents with Artificial Intelligence. powered by
Select sections of text to find additional relevant content using AI-assisted search. powered by
Graphical models are important tools for representing and analyzing uncertain information. Diverse inference methods were developed for efficient computations in these models. In particular, compilation-based inference has recently triggered much research, especially in the probabilistic and the possibilistic frameworks. Even though the inference process follows the same principle in the two frameworks, it depends strongly on the specificity of each of them, namely in the interpretation of handled values (probability\possibility) and appropriate operators (*\min and +\max). This paper emphasizes on common points and unveils differences between the compilation-based inference process in the probabilistic and the possibilistic setting from a spatial viewpoint.