Research on Audit Opinion Prediction of Listed Companies Based on Sparse Principal Component Analysis and Kernel Fuzzy Clustering Algorithm

Table 4

Classification results of different characteristic degradation algorithms.

Dataset	Data set I				Data set II
Method	Cr	F1	G	MCC	Cr	F1	G	MCC
LDA	56.42	67.06	54.71	21.27	63.38	63.89	63.36	26.77
PCA	90.58	90.08	90.44	81.58	88.03	75.86	86.82	77.36
SPCA	93.48	93.23	93.41	87.19	87.32	86.36	87.04	75.40
BSPCA	94.93	94.81	94.90	89.94	90.14	89.23	89.74	81.45
DATASET	Data set III				Data set IV
Method	CR	F1	G	MCC	CR	F1	G	MCC
LDA	51.17	49.80	51.10	2.350	51.94	48.44	51.49	3.910
PCA	83.20	81.39	82.63	67.71	85.16	83.69	84.68	71.50
SPCA	83.59	81.42	82.77	69.11	86.13	85.22	85.91	72.81
BSPCA	83.98	82.10	83.32	69.53	87.10	85.82	86.63	75.43

The bold values are the maximum of the list.