TABLE 4.

Experiment 2 - overall objective and Cohen Kappa metrics for 30 runs.

Method	Best	Worst	Mean	Median	Std	Var
RNN-HASCA	0.034370	0.124386	0.054874	0.124386	2.64E-02	6.99E-04
RNN-SCA	0.047872	0.139525	0.103928	0.124386	4.01E-02	1.61E-03
RNN-GA	0.146481	0.243453	0.204719	0.224223	4.19E-02	1.76E-03
RNN-PSO	0.101064	0.321604	0.222995	0.246318	9.15E-02	8.38E-03
RNN-FA	0.224223	0.386661	0.300600	0.290917	6.67E-02	4.44E-03
RNN-BSO	0.224223	0.332242	0.271686	0.258592	4.51E-02	2.03E-03
RNN-RSA	0.132160	0.319558	0.219585	0.207038	7.70E-02	5.93E-03
RNN-COLSHADE	0.087152	0.224223	0.136661	0.098609	6.21E-02	3.86E-03
Cohen Kappa metrics
RNN-HASCA	0.910288	0.708847	0.863479	0.878882	5.89E-02	3.47E-03
RNN-SCA	0.878882	0.652979	0.746903	0.708847	9.61E-02	9.23E-03
RNN-GA	0.643423	0.208901	0.284108	0.208901	2.68E-01	7.18E-02
RNN-PSO	0.755973	0.167649	0.297728	0.167649	3.34E-01	1.12E-01
RNN-FA	0.245910	−0.028157	0.072584	0.000000	1.23E-01	1.52E-02
RNN-BSO	0.000000	−0.103654	−0.043166	−0.025843	4.41E-02	1.94E-03
RNN-RSA	0.710155	0.411494	0.502563	0.411494	1.47E-01	2.17E-02
RNN-COLSHADE	0.788920	0.000000	0.516435	0.760384	3.65E-01	1.33E-01

In all tables that contain simulation results, the best score in every category is marked bold.