Table 2 Standard evaluation metrics for machine learning reads classification
From: MAC-ErrorReads: machine learning-assisted classifier for filtering erroneous NGS reads
Metric | Definition |
|---|---|
TP | True Positives are error-free reads correctly classified as error-free |
FP | False Positives are erroneous reads incorrectly classified as error-free |
FN | False Negatives are error-free reads incorrectly classified as erroneous |
TN | True Negatives are erroneous reads correctly classified as erroneous |
accuracy | \(\frac{{\left( {TP + TN} \right)}}{{\left( {TP + TN + FP + FN} \right)}}\) |
precision | \(\frac{{\left( {TP} \right)}}{{\left( {TP + FP} \right)}}\) |
recall | \(\frac{{\left( {TP} \right)}}{{\left( {TP + FN} \right)}}\) |
F1-score | \(\frac{{2 \times r{\text{ecall}} \times p{\text{recision}}}}{{\left( {r{\text{ecall}} + p{\text{recision}}} \right)}}\) |
MCC | \(\frac{TP \times TN - FP \times FN}{{\sqrt {\left( {TP + FP} \right) \times \left( {TP + FN} \right) \times \left( {TN + FP} \right) \times \left( {TN + FN} \right)} }}\) |