Characterizations of Decomposable Dependency Models Campos Ibáñez, Luis Miguel Inteligencia artificial Artificial intelligence Decomposable dependency models possess a number of interesting and useful proper- ties. This paper presents new characterizations of decomposable models in terms of in- dependence relationships, which are obtained by adding a single axiom to the well-known set characterizing dependency models that are isomorphic to undirected graphs. We also brie y discuss a potential application of our results to the problem of learning graphical models from data. 2022-11-10T08:15:43Z 2022-11-10T08:15:43Z 1996-12-01 journal article De Campos, L. M. (1996). Characterizations of decomposable dependency models. Journal of Artificial Intelligence Research, 5, 289-300. DOI: [https://doi.org/10.1613/jair.329] https://hdl.handle.net/10481/77868 10.1613/jair.329 eng http://creativecommons.org/licenses/by-nc-nd/4.0/ open access Attribution-NonCommercial-NoDerivatives 4.0 Internacional AI Access Foundation