How can I compare statistically two datasets in RapidMiner

hapaydinhapaydin Member Posts: 4 Contributor I
edited November 2018 in Help

Hi,

I have two data sets. First measured values, second calculated using generate attributes. I would like to compare these data sets statistically. I prefer Leave-one-out Cross Validation [https://en.wikipedia.org/wiki/Cross-validation_(statistics)], but usage of Cross Validation is different in RapidMiner (Divide datasets as training and validation).

Any suggestion?

Data:

3 independent variable( > 200 examples)

1 dependent variable

1 predicted variable

Answers

  • lionelderkrikorlionelderkrikor Moderator, RapidMiner Certified Analyst, Member Posts: 1,195 Unicorn

    Hi @hapaydin,

     

    Can you share your dataset(s) plesase ?

     

    Regards,

     

    Lionel

  • hapaydinhapaydin Member Posts: 4 Contributor I
  • hapaydinhapaydin Member Posts: 4 Contributor I

    csv and jpg of my data

  • lionelderkrikorlionelderkrikor Moderator, RapidMiner Certified Analyst, Member Posts: 1,195 Unicorn

    Hi @hapaydin,

     

    You have directly in RapidMiner access to the staistics of Runoff(M) and Runoff(Pre) in the Statistics

    panel of the Results and then compare their statistics : 

    Rainfall_statistics.png

    or you can use the Charts panels to represent your two datasets (Here an example using histograms) : 

    Rainfall_statistics_2.png

     

    If it's not enough, can you precise what you want .

     

    Regards,

     

    Lionel

    sgenzer
  • lionelderkrikorlionelderkrikor Moderator, RapidMiner Certified Analyst, Member Posts: 1,195 Unicorn

    Hi again @hapaydin,

     

    To go further et to complete my last reply, you can explore the different operators

    of the Statistics Extension (to download and install from the Marketplace).

     

    I hope it will be helpful.

     

    Regards,

     

    Lionel

     

    sgenzer
  • hapaydinhapaydin Member Posts: 4 Contributor I

    Thank you for your kind interst. I will examine Statistics Extension

    sgenzer
Sign In or Register to comment.