n-Grams with a length of 3-6

User: "Kathi"
New Altair Community Member
Updated by Jocelyn
Hi everyone, 

I am currently using the n-Grams operator. If I set the length to 6, all n-Grams will be displayed with a word count of 1-6. I just want to see the n-Grams with a length of 3-6 words. Is that possible?

Thanks!

Find more posts tagged with

Sort by:
1 - 3 of 31
    User: "kayman"
    New Altair Community Member
    Use filter tokens by content, select match and use a regex like ^.*?_.*?_.*
    This will filter on all n-grams having at least 2 underscores, so matching 3 words or more
    User: "lionelderkrikor"
    New Altair Community Member
    Accepted Answer
    Hi @Kathi

    Nice challenge ! 

    Yes, it is possible. You have to duplicate your text processing using : 

    for the first : max length = 6
    for the second : max length = 2

    then Transpose the 2 resulting example sets

    then use a Set Minus operator to keep only the attribute with  3< max length <6

    finally (re)Transpose the final example set.

    The process is in attached file.

    Regards,

    Lionel
    User: "Kathi"
    New Altair Community Member
    OP
    Hi,

    I first tried the set-minus-process, it's working and looks fantastic! Thank you both!

    Regards,
    Kathi