Geological Type Recognition by Machine Learning on In-Situ Data of EPB Tunnel Boring Machines

Table 4

The recognition performance of classifiers in Tianjin Metro Line 9 when the ratio of training and test set is 1 : 9.

	Accuracy (%)	Training time (s)	Test time (s)
KNN	99.8	0.29	81.45
SVM	95.9	3.21	5.60
ANN	97.3	5.76	0.03
CART	99.9	0.03	0.02