Skip to main content

Table 1 Comparison of the performance of various supervised models trained on the training set and evaluated on the validation set

From: Deep semi-supervised learning ensemble framework for classifying co-mentions of human proteins and phenotypes

Method

Precision

Recall

F1

AUROC

SD

PPPred

0.741

0.89

0.809

0.657

N/A

RNN

0.75

0.795

0.772

0.651

0.006

CNN

0.732

0.763

0.747

0.623

0.008

CNN & RNN

0.732

0.822

0.774

0.658

0.005

BERT

0.745

0.936

0.83

0.671

0.005

  1. CNN & RNN is the model using the average prediction probabilities of the individual CNN and RNN models