BMC Bioinformatics

Table 4 Average F1-score results under different sample sizes for pseudo-labeling during fine-tuning based on Tenfold cross-validation

From: meth-SemiCancer: a cancer subtype classification framework via semi-supervised learning utilizing DNA methylation profiles

Sample size	Simulation dataset					Real-studies dataset
Sample size	0	400	600	800	1000	0%	40%	60%	80%	100%
BRCA	0.814	0.820	0.822	0.821	0.825	0.814	0.824	0.822	0.818	0.833
PRAD	0.872	0.882	0.890	0.885	0.899	0.872	0.913	0.901	0.907	0.915
RCC	0.948	0.981	0.979	0.977	0.976	0.948	0.967	0.969	0.974	0.967

Back to article page

ISSN: 1471-2105

Contact us

General enquiries: journalsubmissions@springernature.com