Weighted Decoding for the Competence Reliability Problem of ECOC Multiclass Classification

Algorithm 1

The weight coefficient matrix construction based on SVDD.

	Input: any data-driven ternary matrix , training set , testing sample x
	Training phase
	Step 1: use coding matrix M and train l base classifiers ;
	Step 2: apply SVDD to get the hypersphere (ri, oi) , for each class;
	Testing phase
	Step 3: for a testing instance , compute the distance to each class to get the distance matrix;
	Step 4: use equations (6) and (7) to calculate the weights of base classifiers if M (i, j) = 0;
	Step 5: normalize w;
	Output: base classifier weight vector w