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?

  • 2
    $\begingroup$ What do you mean by "sensitive to classification score"? Sensitivity usually means the true-positive rate, but you use the F1-score. $\endgroup$ – dimpol Oct 31 '16 at 10:32
  • $\begingroup$ I mean Random forest is showing lots of variation in F-score by just changing classification threshold form 0.5 to 0.3. $\endgroup$ – Sangeeta Oct 31 '16 at 16:15

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Browse other questions tagged or ask your own question.