Skip to main content

Table 6 AUC-ROC performance and a 95% CI for the different classification levels for the best standalone and the ensemble models

From: Ensemble of deep learning language models to support the creation of living systematic reviews for the COVID-19 literature

Level

AUC-ROC

BioBERT

Ensemble

Class

91.77 (CI: 90.95, 92.50)

94.33 (CI: 93.70, 94.88)a

Subclass

91.35 (CI: 90.66, 92.01)

94.25 (CI: 93.72, 94.76)a

Sub-subclass

92.06 (CI: 91.56, 92.54)

94.77 (CI: 94.38, 95.12)a

  1. aStatistically significant improvement
Back to article page

Systematic Reviews

ISSN: 2046-4053