Train a Muse
The Train Muse dialog enables you to train a Muse with translation memories and corpora (local ones or offline copies). Muses are local resources, but you can also train offline TMs.
How to begin
Training options section:
•Sensitivity: Use the slider to indicate how many times the expression must occur in the input to be included in the Muse. You can move the slider between 5 and 25, 5 is set as default.
•Precision: Use the slider to indicate how reliable the extracted expressions must be to be included in the Muse. You can move the slider between 0 and 100. If you move the slider towards 0, you get more results but less reliable.
Training data section:
Choose from the Translation memories list the TMs that you would like to include in the Muse. Check the check box next to the TMs to include them. Choose from the Corpora list the corpora that you would like to include in the Muse. Check the check box next to the corpora to include them.
In the What to use from alignments section, you have 2 options:
•Every link, including automatic ones: Choose this option to include every link from alignment, even the non-confirmed ones.
•Only confirmed links: Choose this option to only include confirmed links in the alignment.
Click Start to start the training. When the training finished, memoQ directs you back to the Muses in Project home. Click Cancel to close the dialog without training the Muse.