Learning in Bayesian networks may use a point estimate of the parameters or Bayesian statistics 1 to average over possible model structures and parameters to provide an estimate of the posterior ...