The best approach is to not use stop words at all. That gives better relevance 
with less configuration, so it is a total win.

wunder
Walter Underwood
[email protected]
http://observer.wunderwood.org/  (my blog)

> On Dec 2, 2019, at 12:24 PM, Jörn Franke <[email protected]> wrote:
> 
> You can have different fields by country. I am not sure about your stop words 
> but if they are not occurring in the other languages then you have not a 
> problem. 
> On the other hand: it you need more than stop words (eg lemmatizing, 
> specialized way of tokenization etc) then you need a different field per 
> language. You don’t describe your full use case, but if you have different 
> fields for different language then your client application needs to handle 
> this (not difficult, but you have to be aware).
> Not sure if you need to search a given address in all languages or if you use 
> the language of the user etc.
> 
>> Am 02.12.2019 um 20:13 schrieb yeikel valdes <[email protected]>:
>> 
>> Hi,
>> 
>> 
>> I have an index that stores addresses from different countries.
>> 
>> 
>> As every country has different stop words, I was wondering if it is possible 
>> to apply a different set of stop words depending on the value of a field. 
>> 
>> 
>> Or do I need different indexes/do itnat the ETL step to accomplish this?
>> 
>> 

Reply via email to