Prediction of miRNA-Disease Association Using Deep Collaborative Filtering
Algorithm 1
DCFMDA.
Input: the number of miRNAs, the number of diseases, iterative number , the known miRNA-disease associations , the miRNA similarity matric , the disease similarity .
Output: the predicted score matrix
for each do
for each known miRNA-disease association do
Randomly generate four negative samples;
Obtain the embedding vector of miRNA ;
Obtain the embedding vector of disease ;
Obtain the latent vector of miRNA ;
Obtain the latent vector of disease ;
;
;
;
Concatenating and to form a vector according to formula (21);
Concatenating and to form a vector according to formula (22);
;
Compute by loss function according to formula (24);