Computational Intelligence and Neuroscience

Research Article

A Fast Framework for Abrupt Change Detection Based on Binary Search Trees and Kolmogorov Statistic

Table 3

The summary of simulations on and near the left and right boundaries according to different variance .


Items		Methods
		,				,				,				,
		*BSTKS*	KS		SSA	*BSTKS*	KS		SSA	*BSTKS*	KS		SSA	*BSTKS*	KS		SSA

	Hit rate	*.200*	.055	.036	.160	*.215*	.060	.010	0.0	*.230*	.160	0.0	0.0	*.265*	.165	0.0	.065
	Error	2	20	8	3	3	7	8	12	24	1	134	49	28	21	102	33
	Accuracy	*.937*	.375	.750	.906	*.906*	.781	.750	.625	*.960*	.996	.476	.808	*.891*	.918	.601	.871
	AUC	*.635*	.963	.641	.978	*.656*	.987	.599	.978	*.599*	.946	.780	.797	*.750*	.884	.564	.797

	Hit rate	*.515*	.160	.040	.295	*.485*	.190	.005	0.0	*.490*	.195	0.0	0.0	*.535*	.175	.025	.100
	Error	0	10	4	2	0	3	8	12	1	0	112	10	0	1	90	1
	Accuracy	*1.0*	.687	.875	.937	*1.0*	.906	.750	.625	*.996*	1.0	.562	.960	*1.00*	.996	.648	.996
	AUC	*.831*	.883	.641	.978	*.922*	.927	.599	.978	*.864*	.986	.780	.829	*.999*	.988	.657	.911

	Hit rate	*.655*	.240	0.0	.220	*.725*	.175	0.0	0.0	*.510*	.160	0.0	0.0	*.530*	.150	.005	.065
	Error	0	6	1	2	0	1	4	12	0	0	107	5	0	1	36	4
	Accuracy	*1.00*	.812	.968	.937	*1.0*	.968	.875	.625	*1.0*	1.0	.582	.980	*1.0*	.996	.859	.984
	AUC	*.976*	.979	.770	.978	*.938*	.875	.808	.978	*.978*	.985	.780	.999	*1.0*	.990	.752	.995

	Hit rate	*.715*	210	0.0	.265	*.730*	.215	0.0	0.0	*.530*	.195	0.0	0.0	*.545*	.145	0.0	.060
	Error	0	6	1	2	0	1	1	13	0	0	119	5	0	2	11	4
	Accuracy	*1.0*	.812	.968	.937	*1.0*	.968	.968	.593	*1.0*	1.0	.535	.980	*1.0*	.992	.957	.984
	AUC	*.999*	.960	.770	.978	*.996*	.822	.808	.978	*.999*	.940	.780	.999	*1.0*	.990	.752	.998