Skip to main content

Table 6 Comparison of SVM and StarryNite on the new validation set.

From: Using machine learning to speed up manual image annotation: application to a 3D imaging protocol for measuring single cell gene expression in the developing C. elegans embryo

Metric

CV SVM2

Dev SVM2

StarryNite

True Positives

4306

4571

4874

True Negatives

1273

902

0

False Positives

675

1046

1948

False Negatives

568

303

0

Sensitivity

88.34

93.78

100.0

PPV

86.45

81.38

69.39

Accuracy

81.78

80.23

71.13

  1. Two sets of SVM results are included: from a collection of SVMs trained via cross-validation on the new validation data set ("CV SVM2"), and from a single SVM trained on the new development set ("Dev SVM2"), which is the combination of the initial development and validation sets
Back to article page

BMC Bioinformatics

ISSN: 1471-2105

Contact us