I am working on imbalanced dataset. I am usng three algorithms: RF, SVM and J48. Generally an instance is classified as positive if its classification score is greater than 0.5. However, since I am working on imbalanced data,I perform a small experiment. I compute F-measure of all the classifiers at different classification scores form 0.1 to 0.9. I found that RF is most sensitive to classification score. Does any one have any idea why its is happening?