Evidential networks are frameworks of interest commonly used to represent uncertainty and to reason within the belief function formalism. Despite their success in handling different uncertain situations, the exponential computational complexity which occurs when carrying out the exact inference in these networks makes the use of such models in complex problems difficult. Therefore, with real applications reaching the size of several tens or hundreds of variables, it becomes important to address the serious problem of the feasibility of the exact evidential inference. This paper investigates the issue of applying an approximate algorithm to the belief function propagation in evidential networks. © Springer International Publishing Switzerland 2014.
CITATION STYLE
Laâmari, W., Ben Hariz, N., & Ben Yaghlane, B. (2014). Approximate Inference in Directed Evidential Networks with Conditional Belief Functions Using the Monte Carlo Algorithm. In Communications in Computer and Information Science (Vol. 444 CCIS, pp. 486–497). Springer Verlag. https://doi.org/10.1007/978-3-319-08852-5_50
Mendeley helps you to discover research relevant for your work.