Turkish analyzer

The built-in turkish analyzer can be applied to a text field using the following command:

  1. PUT /turkish-index
  2. {
  3.   "mappings": {
  4.     "properties": {
  5.       "content": {
  6.         "type": "text",
  7.         "analyzer": "turkish"
  8.       }
  9.     }
  10.   }
  11. }

copy

Stem exclusion

You can use stem_exclusion with this language analyzer using the following command:

  1. PUT index_with_stem_exclusion_turkish_analyzer
  2. {
  3.   "settings": {
  4.     "analysis": {
  5.       "analyzer": {
  6.         "stem_exclusion_turkish_analyzer": {
  7.           "type": "turkish",
  8.           "stem_exclusion": ["otorite", "onay"]
  9.         }
  10.       }
  11.     }
  12.   }
  13. }

copy

Turkish analyzer internals

The turkish analyzer is built using the following components:

  • Tokenizer: standard

  • Token filters:

    • apostrophe
    • lowercase (Turkish)
    • stop (Turkish)
    • keyword
    • stemmer (Turkish)

Custom Turkish analyzer

You can create a custom Turkish analyzer using the following command:

  1. PUT /turkish-index
  2. {
  3.   "settings": {
  4.     "analysis": {
  5.       "filter": {
  6.         "turkish_stop": {
  7.           "type": "stop",
  8.           "stopwords": "_turkish_"
  9.         },
  10.         "turkish_stemmer": {
  11.           "type": "stemmer",
  12.           "language": "turkish"
  13.         },
  14.         "turkish_lowercase": {
  15.           "type":       "lowercase",
  16.           "language":   "turkish"
  17.         },
  18.         "turkish_keywords": {
  19.           "type": "keyword_marker",
  20.           "keywords": []
  21.         }
  22.       },
  23.       "analyzer": {
  24.         "turkish_analyzer": {
  25.           "type": "custom",
  26.           "tokenizer": "standard",
  27.           "filter": [
  28.             "apostrophe",
  29.             "turkish_lowercase",
  30.             "turkish_stop",
  31.             "turkish_keywords",
  32.             "turkish_stemmer"
  33.           ]
  34.         }
  35.       }
  36.     }
  37.   },
  38.   "mappings": {
  39.     "properties": {
  40.       "content": {
  41.         "type": "text",
  42.         "analyzer": "turkish_analyzer"
  43.       }
  44.     }
  45.   }
  46. }

copy

Generated tokens

Use the following request to examine the tokens generated using the analyzer:

  1. POST /turkish-index/_analyze
  2. {
  3.   "field": "content",
  4.   "text": "Öğrenciler Türk üniversitelerinde öğrenim görüyor. Numara 123456."
  5. }

copy

The response contains the generated tokens:

  1. {
  2.   "tokens": [
  3.     {"token": "öğrenci","start_offset": 0,"end_offset": 10,"type": "<ALPHANUM>","position": 0},
  4.     {"token": "türk","start_offset": 11,"end_offset": 15,"type": "<ALPHANUM>","position": 1},
  5.     {"token": "üniversite","start_offset": 16,"end_offset": 33,"type": "<ALPHANUM>","position": 2},
  6.     {"token": "öğre","start_offset": 34,"end_offset": 41,"type": "<ALPHANUM>","position": 3},
  7.     {"token": "görüyor","start_offset": 42,"end_offset": 49,"type": "<ALPHANUM>","position": 4},
  8.     {"token": "numar","start_offset": 51,"end_offset": 57,"type": "<ALPHANUM>","position": 5},
  9.     {"token": "123456","start_offset": 58,"end_offset": 64,"type": "<NUM>","position": 6}
  10.   ]
  11. }