Markovian Approach to Time Transition Inference on Bayesian Networks
Ádamo Lima de Santana, Diego L. Cardoso, Carlos Renato, Joo Cost · Sciyo eBooks · 2010
Also known in literature as belief networks, causal networks or probabilistic networks, Bayesian networks (BN) can be seen as models that codify the probabilistic relationships between the variables that represent a given domain (Chen, 2001); being one of the most prominent when considering the easiness of knowledge interpretation achieved. These models possess as components a qualitative (representing the dependencies between the nodes) and a quantitative (conditional probability tables of these nodes) structure, evaluating, in probabilistic terms, these dependencies (Pearl, 1988). Together, these components provide an efficient representation of the joint probability distribution of the variables of a given domain (Russel and Norvig, 2003).