Hi Dan,
Been trying to use edgengram analyzer but i don't understand why it produce
the result in the following order :
1) "DISPLAY_NAME": "Happy Talk"
2) "DISPLAY_NAME": "Happenings",
I suppose the longer the text ("Happy Talk") it should have lower the
scores ?
*The query that i use : *
{ "explain" : true,
"query" : {
"multi_match": {
"query": "*hap*",
"fields": [ "DISPLAY_NAME.AUTOCOMPLETE^6", "PERFORMER.AUTOCOMPLETE" ]
}
},
"sort" : [{
"_score" : { "order" : "desc"}}]
}
*The result :*
https://gist.github.com/cheehoo/10365281
*The mapping :*
"DISPLAY_NAME": {
"type": "multi_field",
"fields": {
"AUTOCOMPLETE": {
"type": "string",
"analyzer": "edge_ngram_keyword_lowercase_analyzer",
"include_in_all": false
},
"NAME": {
"type": "string",
"analyzer": "standard",
"include_in_all": false
}
}
},
"PERFORMER": {
"type": "multi_field",
"fields": {
"AUTOCOMPLETE": {
"type": "string",
"analyzer": "edge_ngram_keyword_lowercase_analyzer",
"include_in_all": false
},
"NAME": {
"type": "string",
"analyzer": "standard",
"include_in_all": false
}
}
},
*And the analyzer settings: *
{
"jdbc_dev": {
"settings": {
"index.analysis.analyzer.string_lowercase.filter": "lowercase",
"index.number_of_replicas": "1",
"index.version.created": "900299",
"index.number_of_shards": "5",
"index.analysis.analyzer.string_lowercase.tokenizer": "keyword",
"index.analysis.tokenizer.my_edge_ngram_tokenizer.token_chars.0": "letter",
"index.analysis.tokenizer.my_edge_ngram_tokenizer.min_gram":
"2",
"index.analysis.tokenizer.my_edge_ngram_tokenizer.token_chars.1": "digit",
"index.analysis.tokenizer.my_edge_ngram_tokenizer.type":
"edgeNGram",
"index.analysis.tokenizer.my_edge_ngram_tokenizer.max_gram":
"5",
"index.analysis.analyzer.edge_ngram_keyword_lowercase_analyzer.filter.1":
"lowercase",
"index.analysis.analyzer.edge_ngram_keyword_lowercase_analyzer.filter.0":
"my_edge_ngram_filter",
"index.analysis.analyzer.edge_ngram_keyword_lowercase_analyzer.tokenizer":
"keyword",
"index.analysis.filter.my_edge_ngram_filter.type": "edgeNGram",
"index.analysis.filter.my_edge_ngram_filter.min_gram": "2",
"index.analysis.filter.my_edge_ngram_filter.max_gram": "30"
}
}
}
Thanks.
On Thu, Apr 10, 2014 at 3:24 AM, Dan Tuffery <[email protected]> wrote:
> You could use edge ngrams:
>
>
> http://www.elasticsearch.org/guide/en/elasticsearch/reference/current/analysis-edgengram-tokenizer.html
>
> Or if you have the correct version of ElasticSearch, you can use the
> Completion Suggester:
>
> http://www.elasticsearch.org/blog/you-complete-me/
>
> Dan
>
>
>
> On Wednesday, April 9, 2014 8:53:21 AM UTC+1, cyrilforce wrote:
>
>> Hi Dan,
>>
>> Good to know that. So what do you suggest if i wanted to have behavior
>> like when i type (search perform in multiple fields eg : DISPLAY_NAME and
>> PERFORMER)
>>
>> *Hel*
>>
>> Then it returns (in the following order) :
>> *Hel*l
>> *Hel*p
>> *Hel*lo
>> *Hel*lo World
>>
>>
>>
>> On Wed, Apr 9, 2014 at 3:45 PM, Dan Tuffery <[email protected]> wrote:
>>
>>> The boost you define in the 'multi_match' query is not being show in the
>>> explain results, so it is not being applied to the score. It should be
>>> displayed in the weight, i.e.
>>>
>>> "description": "weight(DISPLAY_NAME^8:happy in 33593)
>>> [PerFieldSimilarity], result of:"
>>>
>>> The 'phrase_prefix' type is the issue, if you remove that type the boost
>>> will be applied. So it doesn't look like you can combine 'multi_match'
>>> boosting with the 'phrase_prefix' as the type.
>>>
>>> Dan
>>>
>>>
>>> On Tuesday, April 8, 2014 5:15:38 PM UTC+1, cyrilforce wrote:
>>>
>>>> Hi Dan,
>>>>
>>>> I have enabled an analyzer in mapping due to stopword :
>>>>
>>>> "*DISPLAY_NAME*": {
>>>> "type": "string",
>>>> "*analyzer*": "*standard*"
>>>> }
>>>>
>>>>
>>>>
>>>> *The query : *
>>>>
>>>> "multi_match" : {
>>>> "query" : "*happy*",
>>>> "fields" : [ *"DISPLAY_NAME^8*", "*PERFORMER*" ],
>>>> "type": "*phrase_prefix*",
>>>> "operator" : "AND"
>>>> }
>>>>
>>>>
>>>> *The result returned with explain enabled :*
>>>> *https://gist.github.com/cheehoo/10149517
>>>> <https://gist.github.com/cheehoo/10149517>*
>>>>
>>>>
>>>> Thanks
>>>>
>>>>
>>>>
>>>>
>>>>
>>>>
>>>> On Tue, Apr 8, 2014 at 8:44 PM, Dan <[email protected]> wrote:
>>>>
>>>>> Turn on the explain feature to see why example 4 is not getting a
>>>>> higher score.
>>>>>
>>>>> http://www.elasticsearch.org/guide/en/elasticsearch/referenc
>>>>> e/current/search-explain.html
>>>>>
>>>>> I suspect it has something to do with the way you are indexing your
>>>>> data. If you still have issues a gist would help us.
>>>>>
>>>>> Dan
>>>>>
>>>>>
>>>>> On Tuesday, April 8, 2014 1:11:46 PM UTC+1, cyrilforce wrote:
>>>>>>
>>>>>> Hi,
>>>>>>
>>>>>> I have a query as below :
>>>>>>
>>>>>> "multi_match" : {
>>>>>> "query" : "*happy*",
>>>>>> "fields" : [ *"DISPLAY_NAME^8*", "*PERFORMER*" ],
>>>>>> "type": "*phrase_prefix*",
>>>>>> "operator" : "AND"
>>>>>> }
>>>>>>
>>>>>>
>>>>>> Result return in the following order:
>>>>>> 1)
>>>>>> "_score": 2.1704028,
>>>>>> "_source": {
>>>>>> "DISPLAY_NAME": "*Happy*man",
>>>>>>
>>>>>> 2)
>>>>>> "_score": 1.4312989,
>>>>>> "_source": {
>>>>>> "DISPLAY_NAME": "Boishakh (Version 1)",
>>>>>> "PERFORMER": "*Happy*",
>>>>>>
>>>>>> 3)
>>>>>> "_score": 1.2510761,
>>>>>> "_source": {
>>>>>> "DISPLAY_NAME": "Franzl Im Happysound",
>>>>>> "PERFORMER": "Franzl & Die Psayrer",
>>>>>>
>>>>>> 4)
>>>>>> "_score": 1.0920545,
>>>>>> "_source": {
>>>>>> "DISPLAY_NAME": "*Happy*",
>>>>>> "PERFORMER": "Diandra Arjunaidi"
>>>>>>
>>>>>>
>>>>>> As the result shown why the #4 have lesser score computed as i
>>>>>> already added a boost in the "DISPLAY_NAME^6" field. Is that the boost
>>>>>> not
>>>>>> working for multimatch phrase query ?
>>>>>>
>>>>>>
>>>>>>
>>>>>> --
>>>>> You received this message because you are subscribed to a topic in the
>>>>> Google Groups "elasticsearch" group.
>>>>> To unsubscribe from this topic, visit https://groups.google.com/d/to
>>>>> pic/elasticsearch/yEAJ0Ym8PrU/unsubscribe.
>>>>> To unsubscribe from this group and all its topics, send an email to
>>>>> [email protected].
>>>>> To view this discussion on the web visit https://groups.google.com/d/
>>>>> msgid/elasticsearch/bfd38ce9-13b0-4d25-912f-28036992a6df%40goo
>>>>> glegroups.com<https://groups.google.com/d/msgid/elasticsearch/bfd38ce9-13b0-4d25-912f-28036992a6df%40googlegroups.com?utm_medium=email&utm_source=footer>
>>>>> .
>>>>>
>>>>> For more options, visit https://groups.google.com/d/optout.
>>>>>
>>>>
>>>>
>>>>
>>>> --
>>>> Regards,
>>>>
>>>> Chee Hoo
>>>>
>>> --
>>> You received this message because you are subscribed to a topic in the
>>> Google Groups "elasticsearch" group.
>>> To unsubscribe from this topic, visit https://groups.google.com/d/
>>> topic/elasticsearch/yEAJ0Ym8PrU/unsubscribe.
>>> To unsubscribe from this group and all its topics, send an email to
>>> [email protected].
>>> To view this discussion on the web visit https://groups.google.com/d/
>>> msgid/elasticsearch/96218594-90b7-43f4-a0f5-497da059c2f9%
>>> 40googlegroups.com<https://groups.google.com/d/msgid/elasticsearch/96218594-90b7-43f4-a0f5-497da059c2f9%40googlegroups.com?utm_medium=email&utm_source=footer>
>>> .
>>>
>>> For more options, visit https://groups.google.com/d/optout.
>>>
>>
>>
>>
>> --
>> Regards,
>>
>> Chee Hoo
>>
> --
> You received this message because you are subscribed to a topic in the
> Google Groups "elasticsearch" group.
> To unsubscribe from this topic, visit
> https://groups.google.com/d/topic/elasticsearch/yEAJ0Ym8PrU/unsubscribe.
> To unsubscribe from this group and all its topics, send an email to
> [email protected].
> To view this discussion on the web visit
> https://groups.google.com/d/msgid/elasticsearch/608341a2-0d83-48cc-9126-40499553c58b%40googlegroups.com<https://groups.google.com/d/msgid/elasticsearch/608341a2-0d83-48cc-9126-40499553c58b%40googlegroups.com?utm_medium=email&utm_source=footer>
> .
>
> For more options, visit https://groups.google.com/d/optout.
>
--
Regards,
Chee Hoo
--
You received this message because you are subscribed to the Google Groups
"elasticsearch" group.
To unsubscribe from this group and stop receiving emails from it, send an email
to [email protected].
To view this discussion on the web visit
https://groups.google.com/d/msgid/elasticsearch/CAGS0%2Bg8hzpscJ%2Bj8tmZVGuc-mSy6PKT5O5cQYtBwKtjPQ6Or7g%40mail.gmail.com.
For more options, visit https://groups.google.com/d/optout.
