Figure 1
From: A Bayesian network approach to feature selection in mass spectrometry data
Example Bayesian network. A Bayesian network is a directed acyclic graph whose nodes represent random variables. In our work, the root node (A in this figure) always represents the disease state variable, and all other nodes represent the abundance value of specific mass spectrum features. Arcs are assumed to represent causality, so that the state of A causes a change in the probability that B will take on a certain value.