How does SVM perform in imbalanced datasets?

SVM can struggle with imbalanced datasets but can be improved using class weighting.

Without adjustments, SVM may bias toward majority class. Using class weights or adjusting C per class helps penalize misclassification of minority class more heavily.