Skip to main content

Table 5 Performance of mlLGPR, triUMPF and mlXGPR on the multi-organism community dataset CAMI

From: Multi-label classification with XGBoost for metabolic pathway prediction

Metrics and methods

mlLGPR

triUMPF

mlXGPR + RankChain

mlXGPR + RankChain (BioCyc)

Hamming loss (\(\downarrow\))

0.0975

0.0436

0.0482

0.0415

Average Precision (\(\uparrow\))

0.3570

0.7027

0.8366

0.9145

Average Recall (\(\uparrow\))

0.7827

0.5101

0.2657

0.3629

Average F1 score (\(\uparrow\))

0.4866

0.5864

0.4019

0.5185

  1. \(\downarrow\) indicates that a lower score is better while for \(\uparrow\) a higher score is better. The best performing method is bold for each metric. The sample average is calculated for the average precision, recall and F1 score
Back to article page

BMC Bioinformatics

ISSN: 1471-2105

Contact us