Skip to main content

Table 3 The parameter settings of FMSM

From: FMSM: a novel computational model for predicting potential miRNA biomarkers for various human diseases

Parameter

Setting

Size of the known miRNA-disease associations (R)

5430

Number of diseases (nd)

383

Number of miRNA (nm)

495

Regularization weights for latent factor matrices P and Q (β, λ and γ)

Β = λ = γ = 0.1

Maximum number of iterations (T)

100

Regulation weights for average values of R(di, m′) and R(d′, mj) (Wd and Wm)

1

Learning rate (η)

0.01

Sample factor (Æ¿)

3