the "Replace Missing Values(Series)" just didn't work and I can't figure it out

StarfishStarfish Member Posts: 5 Learner I
Hi dear friends,
     this is my example and I am trying to replace the missing value in "位置" which belongs to ID18 
      so I picked out the "Replace Missing Values(Series)" operator and set the parameters like:

          Howerer, RM just kept telling me it couldn't get access to any data because of the wrong attribute type, whereas, I have already made sure that "位置" was regarded as a nominal attribute in RM, which was the exact type of attribute the operator can be applied to.
 
            So here I am very confused about what's going wrong, does this have something to do with the data being in Chinese and not English?


Best regards,
sincerely,
Saefish
Tagged:

Best Answers

  • yyhuangyyhuang Administrator, Employee, RapidMiner Certified Analyst, RapidMiner Certified Expert, Member Posts: 364 RM Data Scientist
    edited October 2021 Solution Accepted
    Thanks for reporting this @Starfish.

    I am trying to reproduce the issue on my sample data, however, it works smoothly with missing Chinese values. Here in the attachment, is my process for your reference.

    FYI, my input data and types


    Please double check the statistics for your input type. From your screenshot, it looks like the date (开放时间,关闭时间) are not loaded correctly....

    Hope this helps..

    YY
  • David_ADavid_A Administrator, Moderator, Employee, RMResearcher, Member Posts: 297 RM Research
    Solution Accepted

    from your screenshots I would guess the process should work as intended.

    I won't exclude that it might have something to do with the Chinese characters, although I would think that it shouldn't be a problem.
    Could you perhaps  try to verify, if it also happens if you have missing values in other columns with Chinese characters?
    Or even share a sample data set with me and I can have a closer look what happens at a code level.

    Best,
    David

Answers

  • tftemmetftemme Administrator, Employee, RapidMiner Certified Analyst, RapidMiner Certified Expert, RMResearcher, Member Posts: 164 RM Research
    Hi @Starfish

    I think you selected the wrong attribute in the attribute parameter. Though I cannot read Chinese characters, I think the second symbol in the attribute name of the highlighted attribute (and also how you spelled it in the text) is different to the one you have in the screenshot of the parameter.

    Though the operator is (correctly) reporting that it cannot find the attribute.

    Best regards,
    Fabian
  • StarfishStarfish Member Posts: 5 Learner I
    thank for your time @yyhuang The Chinese  values do can be read by RM
    and the data(开放时间,关闭时间) are regarded as data_time in my RM, not like the ones in your sample example, which is served as nominal, would this matter?

    Best,
    Starfish

  • StarfishStarfish Member Posts: 5 Learner I
    ohh thank you @David_A, it's so kind of you to say so!

    Actually it turned out to be a small bug I think. Finding that the operator allows text as well, I placed the "Nominal to text" operator in front of "Replace Missing Values(Series)" and started the execution again. What surprised me was that the missing values of “位置”were finally replaced! 

    Then I deleted the "Nominal to text" operator and tried to deal with barely "Replace Missing Values(Series)" again, with the parameter switch to "nominal" type accordingly, the missing values of “位置”were finally replaced. That means I did exactly the same thing 1h ago and after and got different results, so I guess there was a small bug in the operational procedure and somehow it was fixed

    Best,
    Starfish
  • StarfishStarfish Member Posts: 5 Learner I
    thanks for your suggestion@tftemme I checked it up and it was the same attribute~
Sign In or Register to comment.