0410856 20240103182244.120060210235959.9 1-892512-97-1 Las Vegas CSREA Press 2002 7 s. 117-123ArabniaH. R.MunY.PrasadB. uncertain knowledge probability multidimensional distribution cav_un_auth*0101118 Jiroušek Radim UTIA-B Department of Decision Making Theory Ústav teorie informace a automatizace AV ČR, v. v. i. 12A GA201/02/1269 GA ČR cav_un_auth*0005739 KONTAKT 1999-24 AKTION AT CEZ:AV0Z1075907 This paper advocates a probabilistic method for uncertain knowledge representation and processing. In contrast to most other approaches, which are based on graphical Markov modelling (most popular are Bayesian networks, for which the dependence structure is represented by acyclic directed graphs), the described method is rather procedural. It describes a process by which a multidimensional distribution can be composed from pieces of "local knowledge", i.e., from a system of low-dimensional distributions. cav_un_auth*0212930 International Conference on Information and Knowledge Engineering - IKE'02 Las Vegas US 24.06.2002-27.06.2002 BA MTR http://hdl.handle.net/11104/0130943 UTIA-B 20020070 2002 2002 Las Vegas CSREA Press 1-892512-97-1 Proceedings of the International Conference on Information and Knowledge Engineering 117 123 Arabnia H. R. 340 Mun Y. 340 Prasad B. 340