Learning the Bayesian networks (BNs) structure from data has received increasing attention.\nMany heuristic algorithms have been introduced to search for the optimal network that best matches\nthe given training data set. To further improve the performance of ant colony optimization (ACO) in\nlearning the BNs structure, this paper proposes a new improved coevolution ACO (coACO) algorithm,\nwhich uses the pheromone information as the cooperative factor and the differential evolution (DE) as\nthe cooperative strategy. Different from the basic ACO, the coACO divides the entire ant colony into\nvarious sub-colonies (groups), among which DE operators are adopted to implement the cooperative\nevolutionary process. Experimental results demonstrate that the proposed coACO outperforms the\nbasic ACO in learning the BN structure in terms of convergence and accuracy.
Loading....