"Term Clustering, weight to matrix transformation"

hadi_aghasihadi_aghasi Member Posts: 1 Contributor I
edited May 23 in Help
Hi,

I want to detect topics of some Persian  texts. I computed tfidf measure for my extracted  terms by own using .NET programming. I don't know how to do fallowing tasks:

1. convert tfidf values (in database) to matrix form. For clustering of terms (Topic Detection) I need matrix in RapidMiner.
2. how can transpose output matrix as real value for clustering terms instead of documents

Thanks for responding.

Answers

  • landland RapidMiner Certified Analyst, RapidMiner Certified Expert, Member Posts: 2,525   Unicorn
    Hi,
    I'm afraid but I cannot help you. For answering your first question, I would have to know how you did store your data in the database. You have to work this out for yourself or simply load the texts with rapid miner's text plugin.
    There's an operator named "Transpose" which does surprisingly transpose the matrix. You might then apply normal clustering.

    Greetings,
      Sebastian
    neginz
Sign In or Register to comment.