lucene-java-user mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From "merlin.list" <merlin.l...@gmail.com>
Subject Re: lucene double metaphone ranking.
Date Mon, 14 Mar 2011 21:29:20 GMT
On Mar 14, 2011 5:10 PM, Paul Libbrecht wrote:
> Merlin,
>
> the kind of magic such as "prefer an exact match" still has to be programmed.
> Searching in a field with double-metaphone analyzer will only compare tokens by their
double-metaphone-results.
> You probably want query expansion:
>
> text:picasso
>
> to be expanded to:
>
>    text:picasso^3.0 text.stemmed:picass^1.5 text.phonetic:PKS^1.2
>
> paul
>
>
> Le 14 mars 2011 à 22:02, merlin.list a écrit :
>
> Thank you Paul! i shall try your spell.

>> Hi guys,
>>     Here is my noob question:
>>
>> I'm trying to do fuzzy search on first name, last name. I'm using double metaphone
analyzer. and i encountered the following problem.
>> for example, when i search for picasso, "paski" shows up with the same score as the
spelling of "picasso". when i look at the analyzer result of "paksi, picasso" they are both
analyzed as ''PKS". why isn't the exact spelling getting a higher score?
>>
>> thanks.
>
> ---------------------------------------------------------------------
> To unsubscribe, e-mail: java-user-unsubscribe@lucene.apache.org
> For additional commands, e-mail: java-user-help@lucene.apache.org
>


---------------------------------------------------------------------
To unsubscribe, e-mail: java-user-unsubscribe@lucene.apache.org
For additional commands, e-mail: java-user-help@lucene.apache.org


Mime
View raw message