Plot of structural Hamming distance of the networks learned by the sub-optimal learning algorithm from datasets with different sample sizes. This figure plots the SHD of the networks learned by each of the scoring functions for the Breast, Crx, Car, and Diabetes datasets. We display the results for α = 0.5 for BDeu for all datasets because it had the best behavior in terms of SHD.
Liu et al. BMC Bioinformatics 2012 13(Suppl 15):S14 doi:10.1186/1471-2105-13-S15-S14