Multiview Sample Classification Algorithm Based on L1-Graph Domain Adaptation Learning

Algorithm 1

L1-GCMF multiview sample classification algorithm.

Input: data set , , parameters , , .

Output: the target domain classification results

Start:

  to construct the graph , through the principle of sparse representation; to normalize the data sets through

, .

  to initialize , to generate randomly, to obtain through , to obtain through the training of logistic

regression in .

  for   to do

    foreach do

      to update , and through (11)~(12)

      to fix

      to make the norm for each column of and .

      to calculate the objective function through (6)

End