What are the term recognition match value or search depth good for? If I have an Italian term "laboratori" in my text and I set the match value to 70% I would assume that it matches my term "laboratore" in the term base because it differs less than 10%. Isn't is possible to create an algorithm that recognizes all forms of a word (noun, verb), whether it is singular/plural or otherwise declinated? This, of course, would require a different (grammatical) approach to term recognition.
Hi Marlen,
Unknown said:What are the term recognition match value or search depth good for? If I have an Italian term "laboratori" in my text and I set the match value to 70% I would assume that it matches my term "laboratore" in the term base because it differs less than 10%.
This is exactly what it's for. I did a quick test and see this:
Note that "laboratore" is found for "laboratori". Doesn't this work like this for you?
Now, there are limitations with this approach and for agglutinative languages like Hungarian, Japanese or Turkish for example the Multiterm approach has it's limitations. We are going to be trialling a new approach to term recognition that uses a stemming approach similar to what you have described, probably in January. So if you are interested let me know.
In the meantime I'd like to know if you are unable to get the match I showed above?
Regards
Paul
Paul Filkin | RWS Group
________________________
Design your own training!
You've done the courses and still need to go a little further, or still not clear?
Tell us what you need in our Community Solutions Hub
Hi Marlen,
Unknown said:What are the term recognition match value or search depth good for? If I have an Italian term "laboratori" in my text and I set the match value to 70% I would assume that it matches my term "laboratore" in the term base because it differs less than 10%.
This is exactly what it's for. I did a quick test and see this:
Note that "laboratore" is found for "laboratori". Doesn't this work like this for you?
Now, there are limitations with this approach and for agglutinative languages like Hungarian, Japanese or Turkish for example the Multiterm approach has it's limitations. We are going to be trialling a new approach to term recognition that uses a stemming approach similar to what you have described, probably in January. So if you are interested let me know.
In the meantime I'd like to know if you are unable to get the match I showed above?
Regards
Paul
Paul Filkin | RWS Group
________________________
Design your own training!
You've done the courses and still need to go a little further, or still not clear?
Tell us what you need in our Community Solutions Hub
Translate
