Research on a Machine Learning-Based Method for Assessing the Safety State of Historic Buildings

Table 4

Recognition accuracy of RMSE of KCCA-SVM with different dimensions.

Datasets	Dimension	RMSE
		Minimum value	Average value	Maximum value
ORL	10	8.715e − 2	8.927e − 2	9.142e − 2
	15	8.646e − 2	8.893e − 2	9.013e − 2
	20	8.321e − 2	8.581e − 2	8.676e − 2
	25	8.136e − 2	8.327e − 2	8.435e − 2
Yale	10	1.253e − 1	1.277e − 1	1.294e − 1
	15	1.175e − 1	1.263e − 1	1.381e − 1
	20	9.817e − 2	9.924e − 2	1.063e − 1
	25	9.656e − 2	9.721e − 2	9.826e − 2
AR	10	8.486e − 2	8.579e − 2	8.721e − 2
	15	8.101e − 2	8.243e − 2	8.375e − 2
	20	7.836e − 2	7.903e − 2	8.064e − 2
	25	7.811e − 2	7.885e − 2	7.975e − 2
PIE	10	8.721e − 2	8.835e − 2	8.907e − 2
	15	8.367e − 2	8.472e − 2	8.563e − 2
	20	7.916e − 2	8.112e − 2	8.259e − 2
	25	7.875e − 2	8.002e − 2	8.157e − 2