Skip to main content

Table 1 Overlap classification accuracy

From: Improving de novo sequence assembly using machine learning and comparative genomics for overlap correction

Classifier

Accuracy1

False Pos. Rate2

False Neg. Rate3

J48

99.28%

37.7%

0.183%

NaïveBayes

97.82%

12.2%

2.04%

NaïveBayes (-K) 4

99.24%

44.5%

0.127%

Random Forest

99.20%

45.9%

0.149%

UMD Overlapper

74.90%

76.5%

23.3%

  1. 1Accuracy is defined as the number of true positive and true negative overlaps divided by the total number of overlaps
  2. 2The false positive rate is defined as the number of false positive overlaps divided by the number of false overlaps
  3. 3The false negative rate is defined as the number of false negative overlaps divided by the number of true overlaps
  4. 4NaïveBayes with kernel estimation