Details

    • New Feature
    • Status: Closed
    • Major
    • Resolution: Fixed
    • None
    • 4.7, 6.0
    • modules/analysis
    • None
    • New

    Description

      Normalizer+Stemmer+Stopwords for Sorani kurdish (written in the arabic script).

      The most important piece is the normalization: this varies wildly in practice.
      The stemmer is a light stemmer, very simple and not aggressive at all.

      I tested against the pewan test collection, see:

      baseline is StandardAnalyzer.

      short queries (T) TFIDF BM25 I(ne)B2
      baseline 0.2355 0.2473 0.2702
      patch 0.2930 (+24%) 0.3163 (+28%) 0.3309 (+22%)
      long queries (D) TFIDF BM25 I(ne)B2
      baseline 0.3111 0.3185 0.3547
      patch 0.4060 (+31%) 0.4422 (+39%) 0.4800 (+35%)

      Attachments

        1. ckbtestdata.zip
          42 kB
          Robert Muir
        2. LUCENE-5379.patch
          47 kB
          Robert Muir

        Activity

          People

            Unassigned Unassigned
            rcmuir Robert Muir
            Votes:
            0 Vote for this issue
            Watchers:
            1 Start watching this issue

            Dates

              Created:
              Updated:
              Resolved: