Remove correlated attributes with ..... null attributes

lionelderkrikorlionelderkrikor Moderator, RapidMiner Certified Analyst, Member Posts: 1,195 Unicorn
edited December 2018 in Help

Dear all,

 

It is a very special case, however ..... : 

By working on a dataset with many null attributes (all values of these attributes are null), I discover

that Remove Correlated Attributes operator don't take them into account  : They are not removed.

I assume that it is because the definition of the correlation coefficient.

 

What should be the behaviour of this operator in this case ?

 

Regards,

 

Lionel

Tagged:

Best Answer

  • Telcontar120Telcontar120 Moderator, RapidMiner Certified Analyst, RapidMiner Certified Expert, Member Posts: 1,635 Unicorn
    Solution Accepted
    I think "Remove Useless Attributes" should be able to discard the attributes with all or mostly missing values for you.
    You could also use Replace Missing Values with a constant number before the Remove Correlated Attributes approach.
    Brian T.
    Lindon Ventures 
    Data Science Consulting from Certified RapidMiner Experts

Answers

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

    Hi @Telcontar120,

     

    Indeed, Remove Useless Attributes was the solution adapted to my dataset ! (I didn't know this operator).

     

    Thanks you, Brian,

     

    Regards,

     

    Lionel

     

     

Sign In or Register to comment.