I am not sure but I have read that Bagged tree is used to improve the accuracy of a signle tree methods such as C4.5 but applying both of them over the same dataset I got better accuracy with C4.5, 100% while the Bagged tree gave me 97% Is this result acceptable or not?

  • $\begingroup$ How are you getting the 100% and 97%? Is it from testing on the training data? $\endgroup$
    – G5W
    Mar 17 '19 at 14:12
  • $\begingroup$ It is for testing $\endgroup$
    – MakBad
    Nov 29 '19 at 13:52
  • $\begingroup$ Some more details might be helpful. 100% accuracy indicates either a really easy problem, or data leakage (/flawed train-test split), or a small test set. $\endgroup$ Nov 29 '19 at 15:24
  • $\begingroup$ Yes it was a small test data with 400 rows $\endgroup$
    – MakBad
    Nov 29 '19 at 15:28
  • $\begingroup$ I'm still skeptical of 100% accuracy on a test set; 400 is small, but not so small that I'd chalk it up entirely to luck... $\endgroup$ Dec 3 '19 at 2:51

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.