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

Algorithm 1

Kernel Fuzzy C Mean with KNN (KFCM_KNN).

	Input: Given a set of data points , a basis Gaussian kernel function s, the number of clusters , the fuzzy index , the termination criterion and , and the initialization partition matrix .
	Output: The clustering prototypes .
(1)	Procedure KFCM_KNN (Data , Number , kernel functions )
(2)	The partition matrix from FCM as initial membership matrix
(3)	Calculate kernel matrix by .
(4)	Repeat
(5)	Calculate distances by (7)
(6)	Update partition matrix by (6)
(7)	Update clustering prototypes by (5)
(8)	Until or the number of iterations .
(9)	Return and .
(10)	End procedure
(11)	Find the sample closest to the clustering prototypes by the KNN algorithm.