How do I prioritize TMs for a specific translation

I am working on a project using two file based TMs as well as SDL language cloud and MyMemory. Am finding that instead of getting results from my file based TMs, which are obviously preferable, I am getting results from the language cloud and MyMemory. Probably because studio is classifying those results as 100% or CM, and the results from the file based TM's would for the most part be near matches or fuzzy matches.

The problem is that the language cloud and MyMemory results are inaccurate contextually and I would like to change the setting so that any results from the file based TM are prioritized. How do I do that?

Thanks :-)

Parents Reply
  • Hi Ruth

    In addition to putting the TMs in the order of priority, you can also set a Penalty in the same windows as Steve shows above.
    If you apply a penalty of, say 20%, to the 'other' TM sources, then even a 100% or CM match from these others (Cloud, MyMem) will only take priority if your TMs can only produce an 80% match. Just be careful to select a penalty that does not take good hits below the minimum match value in Search settings (next item down in left pane of same window).

    Applying any penalty to these resources will give 100% and CM matches from your TMs priority, the question is to play with the penalty value to get the right balance without excluding useful fuzzy matches from these other resources.

    Best, Simon
Children
No Data