Skip to main content

Table 4 Procedure of the MFNN with P-U learning algorithm

From: Matrix factorization with neural network for predicting circRNA-RBP interactions

Algorithm 2: The MFNN with P-U learning Algorithm

Input:Y: the known interaction matrix,  T: the times of sampling round

Set: Obtain set  P  and  U  from  Y, K: the size of  P  in each sampling round

Output:Fu: unlabeled sample score

Step 1: Initialize uU, t(u) ← 0, MFNN(u) ← 0

Step 2: For  t  from 1 to  T  do

 Randomly sample the set Ut  of size  K  in  U.

 Train a model  MFNNt to discriminate  P  against Ut

 For uU\Ut, update:

MFNN(u) ← MFNN(u) + MFNNt(u)

 t (u) ← t(u) + 1

end For

Step 3: Return  Fu = MFNN(u)/t(u)  for  uU