Research on Model and Algorithm of Multiview and Multilabel Classification Based on Nearest-Neighbor Model

Algorithm 2

Multiview and multilabel optimal chain classifier learning algorithm.

	Input: training set data , the number of classifier chains , and model parameters
	Step 1: Initialize parameter .
	Step 2: The following operations are repeated until the objective function can be minimized.
	For each , first, the corresponding tag sequence is generated according to the algorithm (5), and then, the chain model learning and training are carried out according to the tag sequence by using the dataset from the corresponding perspective; the parameter of the minimized objective function is obtained.
	Output: the whole model framework includes optimal chain classifiers and corresponding weight vectors .