Skip to main content

Table 5 Effects of training data size on prediction performance (AUC) for yeast

From: Protein-protein interaction prediction based on multiple kernels and partial network with linear programming

 

G tt ∼15000

G tt ∼14000

G tt ∼13000

\(\phantom {\dot {i}\!}{RL}_{ {WOLP-K-1}:G_{tn}\sim 5394} \)

0.8658

-

-

\( {RL}_{G_{tn}\sim 7394} \)

0.7931

-

-

\(\phantom {\dot {i}\!} {RL}_{ {EW-K}:G_{tn}\sim 5394} \)

0.7519

-

-

\(\phantom {\dot {i}\!} {RL}_{ {WOLP-K-1}:G_{tn}\sim 5394} \)

-

0.8659

-

\( {RL}_{G_{tn}\sim 8394} \)

-

0.8538

-

\(\phantom {\dot {i}\!} {RL}_{ {EW-K}:G_{tn}\sim 5394} \)

-

0.7537

-

\(\phantom {\dot {i}\!} {RL}_{ {WOLP-K-1}:G_{tn}\sim 5394} \)

-

-

0.8659

\( {RL}_{G_{tn}\sim 9394} \)

-

-

0.8619

\(\phantom {\dot {i}\!} {RL}_{ {EW-K}:G_{tn}\sim 5394} \)

-

-

0.7520