S²NMF: Information Self-Enhancement Self-Supervised Nonnegative Matrix Factorization for Recommendation

Algorithm 2

Optimization algorithm of objective Equation (17).

 Input: S  B;

 Output: ;

 Initialize:N_iter =1, Randomly initialize a nonnegative matrix W and H

 while N_iter <500 do

  For do

   Update according to equation (18): ;

    Update according to equation (19): ;

    Update according to equation (24): ;

   if the convergence of objective function then

      Break;

   N_iter = N_iter +1;

 Return ;