Learning Bayesian network structures by searching for the best ordering with genetic algorithms

In this paper we present a new methodology for inducing Bayesian network structures from a database of cases. The methodology is based on searching for the best ordering of the system variables by means of genetic algorithms. Since this problem of ending an optimal ordering of variables resembles the traveling salesman problem, we use genetic operators that were developed for the latter problem. The quality of a variable ordering is evaluated with the structure-learning algorithm K2. We present empirical results that were obtained with a simulation of the ALARM network.