Skip to main content

Table 5 Performance of the predictive models for each serotype by using independent testing data. PCC was used as the feature selection method. SVM: Support Vector Machine; Sn: Sensitivity; Sp: Specificity; Acc: Accuracy; MCC: Matthews Correlation Coefficient

From: Rapid classification of group B Streptococcus serotypes based on matrix-assisted laser desorption ionization-time of flight mass spectrometry and machine learning techniques

SerotypeBin size of PeaksNumber of featuresClassifiersSnSpAccMCC
I a928Random Forest66.0%61.4%61.9%0.168
SVM19.1%94.9%87.1%0.172
I b330Random Forest55.6%54.8%54.9%0.071
SVM54.0%38.9%41.0%−0.050
III67Random Forest73.0%74.1%73.9%0.405
SVM68.0%71.3%70.6%0.336
V943Random Forest63.6%40.6%43.4%0.028
SVM5.5%94.1%83.4%−0.007
VI718Random Forest70.4%70.3%70.4%0.381
SVM67.6%64.4%65.4%0.297