Re: [Corpora-List] Developing and testing new similarity measures for word clustering

From: Mark P. Line (mark@polymathix.com)
Date: Fri Oct 08 2004 - 22:46:47 MET DST

  • Next message: Dinoj Surendran: "Re: [Corpora-List] Developing and testing new similarity measures for word clustering"

    Normand Peladeau said:
    >
    > I am working now on some modified versions of those indices and I need
    > some ways to benchmark those new similarity measures. I would like to
    > have a series of benchmarks for several kinds of application (dimension
    > reduction, automatic identification of themes, automatic taxonomy
    > development, etc.).

    Although I don't think I can offer any advice on where to get your
    benchmark data without stating the obvious, I do have one suggestion about
    how you might proceed once you know what results you're looking for
    against the benchmark:

    You could use _genetic programming_ to breed indices that have just the
    properties you're looking for (as long as you know what you're looking for
    when you see it).

    -- Mark

    Mark P. Line
    Polymathix
    San Antonio, TX

    This archive was generated by hypermail 2b29 : Fri Oct 08 2004 - 22:51:17 MET DST