French stemmer problem

classic Classic list List threaded Threaded
4 messages Options
Reply | Threaded
Open this post in threaded view
|

French stemmer problem

Renaud Paquay
Hello,

Does anyone know about a modified version of the French Stemmer ?
This one has too many bad results.
For example, if I use the word : "ours" (bear)
The stemmer stemm it into "our".....which doesn't exist in French.
If I have some words like "L'insepecteur" the index process using the
stemmer doesn't work correctly

So the problem is that the results is not accurate

Someone could help ?

Thanks,

Renaud Paquay
Developer and Network Manager
ISIS SA
Rue des Deportes 120
B-4800 Verviers
Tel: +32-(0)87.23.06.90
Fax: +32-(0)87.23.06.54
email : [hidden email]
url : www.isis.be / www.4dbenelux.be


Reply | Threaded
Open this post in threaded view
|

RE: French stemmer problem

Samir Abdou
Hi,

Take a look to http://www.unine.ch/info/clef where you'll find valuable
resources for many languages including French.

Samir  


-----Message d'origine-----
De : Renaud Paquay [mailto:[hidden email]]
Envoyé : vendredi, 22. décembre 2006 10:54
À : [hidden email]
Objet : French stemmer problem

Hello,

Does anyone know about a modified version of the French Stemmer ?
This one has too many bad results.
For example, if I use the word : "ours" (bear)
The stemmer stemm it into "our".....which doesn't exist in French.
If I have some words like "L'insepecteur" the index process using the
stemmer doesn't work correctly

So the problem is that the results is not accurate

Someone could help ?

Thanks,

Renaud Paquay
Developer and Network Manager
ISIS SA
Rue des Deportes 120
B-4800 Verviers
Tel: +32-(0)87.23.06.90
Fax: +32-(0)87.23.06.54
email : [hidden email]
url : www.isis.be / www.4dbenelux.be




---------------------------------------------------------------------
To unsubscribe, e-mail: [hidden email]
For additional commands, e-mail: [hidden email]

Reply | Threaded
Open this post in threaded view
|

Re: French stemmer problem

Mark Miller-3
In reply to this post by Renaud Paquay
Non of the stemmers always stem to a valid word. It is not important as
you should be stemming the query as well. The only thing that is
important is that each word always stems to the same base. Many English
words do not stem to real English words with the English stemmer either.

Renaud Paquay wrote:

> Hello,
>
> Does anyone know about a modified version of the French Stemmer ?
> This one has too many bad results.
> For example, if I use the word : "ours" (bear)
> The stemmer stemm it into "our".....which doesn't exist in French.
> If I have some words like "L'insepecteur" the index process using the
> stemmer doesn't work correctly
>
> So the problem is that the results is not accurate
>
> Someone could help ?
>
> Thanks,
>
> Renaud Paquay
> Developer and Network Manager
> ISIS SA
> Rue des Deportes 120
> B-4800 Verviers
> Tel: +32-(0)87.23.06.90
> Fax: +32-(0)87.23.06.54
> email : [hidden email]
> url : www.isis.be / www.4dbenelux.be
>
>
>
>  

---------------------------------------------------------------------
To unsubscribe, e-mail: [hidden email]
For additional commands, e-mail: [hidden email]

Reply | Threaded
Open this post in threaded view
|

Re: French stemmer problem

Patrek
Hi Renaud,

Maybe you should take a look at the Morphalou project (
http://actarus.atilf.fr/lexiques/morphalou/) it is a database of lemma and
forms in French.

You could extract the data and create a synonym index or something.

Don't hesitate to contact me off list (and in French if needed) for more
info.

Patrick

On 12/22/06, Mark Miller <[hidden email]> wrote:

>
> Non of the stemmers always stem to a valid word. It is not important as
> you should be stemming the query as well. The only thing that is
> important is that each word always stems to the same base. Many English
> words do not stem to real English words with the English stemmer either.
>
> Renaud Paquay wrote:
> > Hello,
> >
> > Does anyone know about a modified version of the French Stemmer ?
> > This one has too many bad results.
> > For example, if I use the word : "ours" (bear)
> > The stemmer stemm it into "our".....which doesn't exist in French.
> > If I have some words like "L'insepecteur" the index process using the
> > stemmer doesn't work correctly
> >
> > So the problem is that the results is not accurate
> >
> > Someone could help ?
> >
> > Thanks,
> >
> > Renaud Paquay
> > Developer and Network Manager
> > ISIS SA
> > Rue des Deportes 120
> > B-4800 Verviers
> > Tel: +32-(0)87.23.06.90
> > Fax: +32-(0)87.23.06.54
> > email : [hidden email]
> > url : www.isis.be / www.4dbenelux.be
> >
> >
> >
> >
>
> ---------------------------------------------------------------------
> To unsubscribe, e-mail: [hidden email]
> For additional commands, e-mail: [hidden email]
>
>