Table 2 Comparison of the performance of the four machine learning models for gene prioritization.

From: A gene prioritization method based on a swine multi-omics knowledgebase and a deep learning model

Models

Accuracy

Precision

Recall

F1-Measure

LR

0.657

0.686

0.571

0.623

LinearSVC

0.639

0.658

0.571

0.612

MLP

0.692

0.678

0.726

0.701

CNN

0.734

0.729

0.738

0.734

  1. The F1- Measure of the two linear models (LR and LinearSVC) with strong explanatory power were lower than deep learning models (MLP and CNN) that were based on neural networks, which suggested the deep learning models were superior to the linear models. Between MLP and CNN, the accuracy, precision, and F1- Measure of CNN were higher than MLP, and the performance of CNN was slightly better than MLP.
  2. LR Logistic regression, LinearSVC Linear Support Vector Classifier, MLP Multi-Layer Perceptron, CNN Convolutional Neural Networks.
Back to article page