Numbers in memoQ
Posted by Péter Botta on 31 October 2013 08:31 AM
Numbers in memoQ
In memoQ number-handling comes up in several contexts.
In Project home> Settings > Auto-translatables settings there are some built-in number-handling rules for certain languages, but you can also specify your own rules and patterns with regular expressions. These rules are applied to your source text, and elements that are recognized will appear as auto-translatable hits in the results pane in green (default), and the hits in the source text are also highlighted in green. These hits appear in the Autopick list if you press the Ctrl key.
2. Autopick number recognizer
The numbers that appear in the Autopick list can come from the auto-translatable hits, but there is a completely separate number recognizer in memoQ that looks for numbers in the source text and converts them automatically, according to your settings. You don’t have to add rules for this function to work, it will recognize numbers immediately. If you see a number repeated in the Autopick list, it means that it was recognized by an auto-translatable rule, but also by the number recognizer.
3. Populate number only segments
If you have a lot of segments that contain only numbers (and perhaps some non-alphabetical characters), you can fill these segments automatically.
1. Go to Operations (in memoQ 2014 and higher: on the Preparation ribbon tab) > Populate Number-Only Segments
2. Choose your settings
You can choose the scope where you want to fill your segments.
If you check Segments with multiple numbers those segments that have several numbers (separated by space(s)) will also be filled, and not only the segments with one single number.
If you check Non-alphabetical characters or tags besides numbers those segments that have other characters (only non-alphabetical) as well as numbers will also be filled, together with these extra characters. (This is useful eg. for formulas, phone numbers.)
If you check Populate unconfirmed rows with text in target then your current, unconfirmed segments will be overwritten.
You can also set the status of your segments.
3. When you click OK your segments will be filled with the transformed numbers. (More on transformation later.)
4. Fragment assembly
In the fragment assembly hits (purple ones) a complete segment is made up of the hits coming from different sources, and the original source text where there are no hits. Numbers are one kind of hits that are replaced in the segment. The numbers are found with the same recognizer as in Autopick or in the “Populate Number-Only Segments” function, and are also transformed according to your settings.
The settings are on the second tab in Translation > Predictive Typing and Autopick.
5. Quality assurance
The QA module in memoQ checks numbers: their format according to specific language settings, and whether the numbers in the source and the target correspond to each other.
Currently the algorithm to find numbers in the text is different from the previously mentioned cases. It is somewhat simpler, and it doesn’t take your Autopick settings into consideration, but the QA module has different settings.
The settings are in Project home > Settings > QA Settings > Numbers tab. You can choose what you want to be checked, and you can also add custom number formats for different languages.
The built-in number recognizer
The built-in number recognizer is used with some of the previously mentioned functionality in memoQ from version 2013:
All these functionalities use the same number recognizer with the same settings. The settings are available from the Translation Settings on the Translation ribbon tab > Predictive Typing And Autopick on the Autopick tab (the bottom part).
At the bottom of the window you can choose your settings. These settings will influence how the number recognizer will transform your numbers in the target segment. You can: