OpenSearch/docs/reference/analysis/tokenizers/ngram-tokenizer.asciidoc
Christoph Büscher 25aac4f77f
Remove include_type_name in asciidoc where possible (#37568)
The "include_type_name" parameter was temporarily introduced in #37285 to facilitate
moving the default parameter setting to "false" in many places in the documentation
code snippets. Most of the places can simply be reverted without causing errors.
In this change I looked for asciidoc files that contained the
"include_type_name=true" addition when creating new indices but didn't look
likey they made use of the "_doc" type for mappings. This is mostly the case
e.g. in the analysis docs where index creating often only contains settings. I
manually corrected the use of types in some places where the docs still used an
explicit type name and not the dummy "_doc" type.
2019-01-18 09:34:11 +01:00

306 lines
6.1 KiB
Plaintext

[[analysis-ngram-tokenizer]]
=== NGram Tokenizer
The `ngram` tokenizer first breaks text down into words whenever it encounters
one of a list of specified characters, then it emits
https://en.wikipedia.org/wiki/N-gram[N-grams] of each word of the specified
length.
N-grams are like a sliding window that moves across the word - a continuous
sequence of characters of the specified length. They are useful for querying
languages that don't use spaces or that have long compound words, like German.
[float]
=== Example output
With the default settings, the `ngram` tokenizer treats the initial text as a
single token and produces N-grams with minimum length `1` and maximum length
`2`:
[source,js]
---------------------------
POST _analyze
{
"tokenizer": "ngram",
"text": "Quick Fox"
}
---------------------------
// CONSOLE
/////////////////////
[source,js]
----------------------------
{
"tokens": [
{
"token": "Q",
"start_offset": 0,
"end_offset": 1,
"type": "word",
"position": 0
},
{
"token": "Qu",
"start_offset": 0,
"end_offset": 2,
"type": "word",
"position": 1
},
{
"token": "u",
"start_offset": 1,
"end_offset": 2,
"type": "word",
"position": 2
},
{
"token": "ui",
"start_offset": 1,
"end_offset": 3,
"type": "word",
"position": 3
},
{
"token": "i",
"start_offset": 2,
"end_offset": 3,
"type": "word",
"position": 4
},
{
"token": "ic",
"start_offset": 2,
"end_offset": 4,
"type": "word",
"position": 5
},
{
"token": "c",
"start_offset": 3,
"end_offset": 4,
"type": "word",
"position": 6
},
{
"token": "ck",
"start_offset": 3,
"end_offset": 5,
"type": "word",
"position": 7
},
{
"token": "k",
"start_offset": 4,
"end_offset": 5,
"type": "word",
"position": 8
},
{
"token": "k ",
"start_offset": 4,
"end_offset": 6,
"type": "word",
"position": 9
},
{
"token": " ",
"start_offset": 5,
"end_offset": 6,
"type": "word",
"position": 10
},
{
"token": " F",
"start_offset": 5,
"end_offset": 7,
"type": "word",
"position": 11
},
{
"token": "F",
"start_offset": 6,
"end_offset": 7,
"type": "word",
"position": 12
},
{
"token": "Fo",
"start_offset": 6,
"end_offset": 8,
"type": "word",
"position": 13
},
{
"token": "o",
"start_offset": 7,
"end_offset": 8,
"type": "word",
"position": 14
},
{
"token": "ox",
"start_offset": 7,
"end_offset": 9,
"type": "word",
"position": 15
},
{
"token": "x",
"start_offset": 8,
"end_offset": 9,
"type": "word",
"position": 16
}
]
}
----------------------------
// TESTRESPONSE
/////////////////////
The above sentence would produce the following terms:
[source,text]
---------------------------
[ Q, Qu, u, ui, i, ic, c, ck, k, "k ", " ", " F", F, Fo, o, ox, x ]
---------------------------
[float]
=== Configuration
The `ngram` tokenizer accepts the following parameters:
[horizontal]
`min_gram`::
Minimum length of characters in a gram. Defaults to `1`.
`max_gram`::
Maximum length of characters in a gram. Defaults to `2`.
`token_chars`::
Character classes that should be included in a token. Elasticsearch
will split on characters that don't belong to the classes specified.
Defaults to `[]` (keep all characters).
+
Character classes may be any of the following:
+
* `letter` -- for example `a`, `b`, `ï` or `京`
* `digit` -- for example `3` or `7`
* `whitespace` -- for example `" "` or `"\n"`
* `punctuation` -- for example `!` or `"`
* `symbol` -- for example `$` or `√`
TIP: It usually makes sense to set `min_gram` and `max_gram` to the same
value. The smaller the length, the more documents will match but the lower
the quality of the matches. The longer the length, the more specific the
matches. A tri-gram (length `3`) is a good place to start.
The index level setting `index.max_ngram_diff` controls the maximum allowed
difference between `max_gram` and `min_gram`.
[float]
=== Example configuration
In this example, we configure the `ngram` tokenizer to treat letters and
digits as tokens, and to produce tri-grams (grams of length `3`):
[source,js]
----------------------------
PUT my_index
{
"settings": {
"analysis": {
"analyzer": {
"my_analyzer": {
"tokenizer": "my_tokenizer"
}
},
"tokenizer": {
"my_tokenizer": {
"type": "ngram",
"min_gram": 3,
"max_gram": 3,
"token_chars": [
"letter",
"digit"
]
}
}
}
}
}
POST my_index/_analyze
{
"analyzer": "my_analyzer",
"text": "2 Quick Foxes."
}
----------------------------
// CONSOLE
/////////////////////
[source,js]
----------------------------
{
"tokens": [
{
"token": "Qui",
"start_offset": 2,
"end_offset": 5,
"type": "word",
"position": 0
},
{
"token": "uic",
"start_offset": 3,
"end_offset": 6,
"type": "word",
"position": 1
},
{
"token": "ick",
"start_offset": 4,
"end_offset": 7,
"type": "word",
"position": 2
},
{
"token": "Fox",
"start_offset": 8,
"end_offset": 11,
"type": "word",
"position": 3
},
{
"token": "oxe",
"start_offset": 9,
"end_offset": 12,
"type": "word",
"position": 4
},
{
"token": "xes",
"start_offset": 10,
"end_offset": 13,
"type": "word",
"position": 5
}
]
}
----------------------------
// TESTRESPONSE
/////////////////////
The above example produces the following terms:
[source,text]
---------------------------
[ Qui, uic, ick, Fox, oxe, xes ]
---------------------------