LibSVM classification results differ between Weka and Rapidminer

MsinMsin Member Posts: 1 Newbie
Hi,

Using Weka and Rapidminer, I got different classification results on the same dataset with LibSVM. Even when I made the same parameters that I could change on RapidMiner in both programs, Weka gave 75% and Rapidminer 55% classification accuracy.

Also, RapidMiner SVM and Weka SMO results differ greatly too.

I would be very grateful if you could explain the reasons for these two problems.

Thanks.

Answers

  • MartinLiebigMartinLiebig Administrator, Moderator, Employee, RapidMiner Certified Analyst, RapidMiner Certified Expert, University Professor Posts: 3,485 RM Data Scientist
    Well, its kind of expected. Slight differences in implementation choices can make a big impact.

    But also make sure that you set all random seeds to the same values to make fair comparisons.

    BR,
    Martin
    - Sr. Director Data Solutions, Altair RapidMiner -
    Dortmund, Germany
Sign In or Register to comment.