Thanks for the reply.
It would be cool to have the type in the index. Imagine if you had different
types like
person, place, event or even subject, predicate, object. It would greatly
enhance the search capabilities of lucene.
Andrew
-----Original Message-----
From: Erik Hatcher <[EMAIL PROTECTED]>
Sent: Jul 8, 2005 11:01 AM
To: java-user@lucene.apache.org
Subject: Re: How to get the un-stemed word
On Jul 8, 2005, at 9:08 AM, Andrew Boyd wrote:
> Hi all,
> I am using the snowball stemmer and for all my searches that
> works fine.
> However, I have a need to display the un-stemmed word after doing
> some term vector analysis.
>
> I was thinking that I might insert the real word at the same
> position as the stemed word but give
> the real word a type of say "r-word". My question is how can I get
> the r-word from the TermVectors?
>
> If there is not a way to get it from the TermVector is there a way
> to search/retrieve by token type?
Currently token type is not stored in the index - it is used solely
during analysis but does not go any further.
It would be quite cool for that type to carry into the index and be
surfaced for searching.
You will need to develop another approach, perhaps using two
different indexes, or two different fields.
Erik
---------------------------------------------------------------------
To unsubscribe, e-mail: [EMAIL PROTECTED]
For additional commands, e-mail: [EMAIL PROTECTED]
---------------------------------------------------------------------
To unsubscribe, e-mail: [EMAIL PROTECTED]
For additional commands, e-mail: [EMAIL PROTECTED]
