Fig. 3From: Automated classification and characterization of the mitotic spindle following knockdown of a mitosis-related proteinROC plot and confusion matrix for predictions of spindle class by a SVM learning algorithm. a ROC plot showing the output of the used learning algorithm. SVM generated the largest Area Under the Curve (AUC) with a value of 0.92 and a true positive rate of 0.85 as marked by the point (0.17, 0.87). b Confusion matrix: the value shown inside the green box is derived from a true prediction while the number inside the grey box shows a false predictionBack to article page