Does MT have a problem with accented characters?

Does MT have a problem with accented characters?

 

My TB has entries for “Einträge” (with umlaut) and “eintragen” (without umlaut)

 

The matched term is incorrectly “eintragen” rather than “Einträge” (which does not appear in the term match list)

 

Parents
  • There does seem to be a problem here and I'm not sure why.  I'll have to work with the support team and come back to you.  But in the meantime I tested importing my MultiTerm termbase into the cloud and in there it works well:

    The linguistic search in Language Cloud is much improved over MultiTerm, so with the defaults it works fine:

    Perhaps you could try working with the cloud as well?

Reply
  • There does seem to be a problem here and I'm not sure why.  I'll have to work with the support team and come back to you.  But in the meantime I tested importing my MultiTerm termbase into the cloud and in there it works well:

    The linguistic search in Language Cloud is much improved over MultiTerm, so with the defaults it works fine:

    Perhaps you could try working with the cloud as well?

Children
  • Thanks Paul. I will look into the Cloud solution, although because I do not always have an internet connection, an upgraded integrated term recognition would be better (it is an important QA tool). Some of the false positives should be easy to solve, such as the following:

    “service type definition” OK, but “service type definitions” NOK for 60% match level