Selection and replacement problem

j_quigleyj_quigley Member Posts: 8 Contributor I
edited December 2018 in Help

Hi all,

 

The problem I am trying to solve is as follows.

 

I have an attribute (A1) with a range of values [4rf, 56X, 78t...]

The values of (A1)  are also attributes in themselves in the same example set. (A4rf),  (A56X),  (A78t).

For each row I want to select the corresponding attribute to the one in (A1) and replace the value of the corrsponding attribute with a given value.

 

I found a solution to doing this with lots of filters and the replace operator but it seems like a lon winded and very un-elegant solution.

 

Do you have a better way to do this?

 

Many thanks,

 

Jamie

Answers

  • Telcontar120Telcontar120 Moderator, RapidMiner Certified Analyst, RapidMiner Certified Expert, Member Posts: 1,635 Unicorn
    Do you want to post your existing solution process and/or a sample data file for others to use as a starting point? That is usually helpful for other community members.
    My initial diagnosis is that a Join to create a new attribute from the original values via a lookup table, followed by a Pivot, should do the trick, but a sample data file will be helpful to demonstrate.
    Brian T.
    Lindon Ventures 
    Data Science Consulting from Certified RapidMiner Experts
  • j_quigleyj_quigley Member Posts: 8 Contributor I

    Hi Brian,

     

    Many thanks for responding - 

    Creating a new attribute isn't necessary, it already exists. I wish to populate it with a spcified value for the given clause below.

     

    Here is a conceptual view of my data. (although the total number of attributes is 120)

     

    A1 - A2 - A3 - A4

    x1 -  y1 - z1 - A1

    x2 -  y6 - z4 - A1

    x7 - y3 - z5 - A2

    x4 - y4 - z4 - A3

     

    where A4 = value An 

    replace value of attribute An with a given value.

     

    So if the given value is "sheepskin" the data would look like this:

     

    A1 -              A2 -              A3 -              A4

    sheepskin -  y1 -               z1 -               A1

    sheepskin -  y6 -               z4 -               A1

    x7 -              sheepskin -   z5 -               A2

    x4 -               y4 -              sheepskin -   A3

     

    Hope that makes things clearer!

  • SGolbertSGolbert RapidMiner Certified Analyst, Member Posts: 344 Unicorn

    I'm afraid you lost me with the example :(

     

    If I understood correctly, then first attribute has as values the names of the attributes to replace. For example:

     

     x   A1   A2   A3  A4
    A1   x      y     z    p
    A2   a      b    c    d

     

    with given value of xXx

    will turn to

     

    x     A1   A2    A3  A4
    A1  xXx    y       z     p
    A2   a     xXx     c    d

  • Telcontar120Telcontar120 Moderator, RapidMiner Certified Analyst, RapidMiner Certified Expert, Member Posts: 1,635 Unicorn

    Sorry, you have definitely lost me here.  

    Brian T.
    Lindon Ventures 
    Data Science Consulting from Certified RapidMiner Experts
Sign In or Register to comment.