Skip to main content

Table 2 The performance of OptNCMiner and baseline models with the base dataset and transfer learning dataset

From: OptNCMiner: a deep learning approach for the discovery of natural compounds modulating disease-specific multi-targets

Model

Performance metric1

Base dataset

Transfer learning dataset

OptNCMiner

Recall

0.833

0.871

AUROC

0.632

0.787

Accuracy

0.440

0.713

Cosine similarity

Recall

0.573

0.696

AUROC

0.643

0.761

Accuracy

0.708

0.818

Naïve bayes classifier

Recall

0.322

0.483

AUROC

0.623

0.696

Accuracy

0.909

0.887

Logistic regression

Recall

0.212

0.581

AUROC

0.606

0.785

Accuracy

0.978

0.969

Random forest

Recall

0.677

0.479

AUROC

0.343

0.241

Accuracy

0.028

0.027

Multi-layer perceptron

Recall

0.361

0.824

AUROC

0.676

0.818

Accuracy

0.972

0.899

  1. 1All performance metrics are weighted averages of the results of all proteins comprising the dataset