Add new column if then in existing column

syazwaansyazwaan MemberPosts:3Newbie
I have a dataset something like this:

I want to make a new column if data in column_1 is start with a number and ends with a letter and return 'Complete' or 'Incomplete' in a new column.

Expected output:

Thank you.
Tagged:

Best Answer

  • BalazsBaranyBalazsBarany Administrator, Moderator, Employee, RapidMiner Certified Analyst, RapidMiner Certified ExpertPosts:824Unicorn
    Solution Accepted
    Hi!

    You could use Generate Attributes for this.
    Determining if a text starts with a number and ends with a letter is the domain of regular expressions. These are usable in different places in RapidMiner, there's even a nice regular expression editor for example in the Replace operator.

    In Generate Attributes, you would use theif()andmatches()functions. Here's an example for your use case:

    if(matches(column_1, "^[0-9]+.+[a-zA-Z]$"), "Complete", "Incomplete")

    The regexp is a bit terrifying in the beginning but it is simple to explain.

    ^ beginning of the string
    [0-9] character class: Numbers from 0 to 9
    + at least one of the element before (the character class)
    .+ anything
    [a-zA-Z] letters (if you need additional characters, enter them after the Z)
    美元的字符串

    So the logic goes: If the value of column_1 matches the regular expression, theiffunction returns the second argument (Complete), else the third one.

    Regards,
    Balázs

Answers

  • Telcontar120Telcontar120 Moderator, RapidMiner Certified Analyst, RapidMiner Certified Expert, MemberPosts:1,630Unicorn
    You could also use the Generate Copy operator and then use Generate Attributes with PREFIX and SUFFIX functions if you simply want to keep separate attributes for both the first and last character of the raw data (this may be helpful depending on what else you want to do with it or if those codes have some meaning).
    Brian T.
    Lindon Ventures
    Data Science Consulting from Certified RapidMiner Experts
    BalazsBarany
Sign InorRegisterto comment.