Skip to main content

Table 3 Statistics of binary classification models built upon different feature sets and using different machine learning algorithms

From: Predicting drug target interactions using meta-path-based semantic network analysis

topological feature

Dataset

Random Forest

Support Vector Machine

mtry

F1-score

C

λ

F1-score

Feature set I

Training

12

0.780

8

0.250

0.766

Test

0.735

0.719

Feature set II

Training

13

0.844

16

0.062

0.810

Test

0.790

0.763

Feature set III

Training

13

0.859

16

0.016

0.843

Test

0.810

0.798