Bayesian Network with Association Rules Applied in the Recognition of Handwritten Digits
Classification Based on Association (CBA) algorithm built a classifier based on the association rules, but without considering the uncertainty in the classification problem. This paper proposed a Bayesian network classifier based on the association rules. The algorithm extracts the candidate set uses association rules and classification algorithms related to the network, then uses “greedy hill-climbing algorithm” to learn network structure to get a better topology, and verify that this algorithm is valid on handwritten numeral recognition.
W. Q. Zhao et al., "Bayesian Network with Association Rules Applied in the Recognition of Handwritten Digits", Advanced Materials Research, Vol. 187, pp. 7-12, 2011