[Egothor-tech] posting info and stemming

[Leo Galambos]

Peter Halacsy wrote:

> One more question. At indexing I'd like to store the stemmed and 
> original form of a word on the same position. Suppose my stemmer stems 
> the word "goes" to "go".
>
> the input is :
>
> as far as it goes -> as far as it goes/go
>
> The query would be stemmed the same way. You search with: "goes" The 
> sentence above should be ranked higher than the others containing the 
> word "go"
>
> Why to store in the same position? To be able to handle phrase and 
> proximity searches as well.
>
> I know I can do it: I should set the token's reloffset to zero if the 
> token is a stemmed form of an original?


Hi!

Yes, if you set Token::reloffset to 0, it will do the job.

Cheers,
Leo

-- 
Leo Galambos
Faculty of Mathematics and Physics, DSE
Malostranske namesti 25
Prague 1
CZE

http://kocour.ms.mff.cuni.cz/~galambos/
http://www.egothor.org/Main/LeoGalambos