Ant Colony Optimization with Three Stages for Independent Test Cost Attribute Reduction

Table 5

Results of 𝜆-information gain algorithm and the test cost based ant colony optimization with centralized deletion and distributed deletion.

Dataset	Distribution	Information gain			GA-1			GA-2			ACO centralized			ACO distributed
		FOF	MEF	AEF	FOF	MEF	AEF	FOF	MEF	AEF	FOF	MEF	AEF	FOF	MEF	AEF
	Uniform	0.915	0.193	0.004	0.966	0.118	0.001	0.947	0.224	0.002	0.973	0.177	0.001	0.975	0.216	0.001
Zoo	Normal	0.833	0.028	0.001	0.970	0.109	0.001	0.924	0.101	0.002	0.849	0.040	0.002	0.844	0.048	0.002
	Pareto	0.999	0.167	0.000	0.998	0.167	0.000	0.993	0.200	0.001	1.000	0.000	0.000	1.000	0.000	0.000
	Uniform	0.998	0.053	0.000	1.000	0.000	0.000	0.999	0.009	0.000	0.997	0.065	0.000	1.000	0.000	0.000
Voting	Normal	1.000	0.000	0.000	1.000	0.000	0.000	1.000	0.000	0.000	1.000	0.000	0.000	1.000	0.000	0.000
	Pareto	1.000	0.000	0.000	1.000	0.000	0.000	0.995	0.067	0.000	1.000	0.000	0.000	1.000	0.000	0.000
	Uniform	0.877	0.055	0.002	1.000	0.000	0.000	1.000	0.000	0.000	1.000	0.000	0.000	1.000	0.000	0.000
Tic-tac-toe	Normal	0.408	0.018	0.003	1.000	0.000	0.000	1.000	0.000	0.000	1.000	0.000	0.000	1.000	0.000	0.000
	Pareto	0.986	0.125	0.002	1.000	0.000	0.000	1.000	0.000	0.000	1.000	0.000	0.000	1.000	0.000	0.000
	Uniform	0.762	0.523	0.023	0.448	1.000	0.067	0.766	0.543	0.019	0.948	0.085	0.002	0.946	0.144	0.002
Mushroom	Normal	0.176	0.306	0.010	0.321	0.401	0.063	0.551	0.340	0.022	0.958	0.020	0.000	0.966	0.020	0.000
	Pareto	0.733	0.500	0.064	0.935	0.400	0.015	0.949	0.500	0.012	1.000	0.000	0.000	1.000	0.000	0.000