WorldServer allows custom stemming algorithms to be implemented and
applied to translation memories.
Word stemmers impact the search process in that they impact how the
translation memory identifies word units during the search process
(and how it stores search information when storing new entries). Stemming
allows for various forms of words to be normalized to a simpler form,
and this simpler form can be used to match to the more specific forms.
For instance, without stemming, searching for "cat" would not be able
to yield a fuzzy match to "cats."
To add a stemmer component
-
From , choose
Word Stemmer from the
Custom component type drop-down menu.
-
Click
Add.
WorldServer displays the
Add Custom Word Stemmer Component dialog.
-
Click
Browse. Use the
Choose
file dialog box to navigate your local directory structure
and select the ZIP file containing the stemmer.
-
Click
Open.
The
Choose file dialog box closes and the selected files is listed in the
Add Custom Word Stemmer Component dialog.
-
Click
OK.
WorldServer returns to the
Customization page.
Please refer to the
WorldServer Translation Memory Administration
Guide (TMAG) and the SDK documentation for more information.