# Key terms, understanding MemoQ TB morphology

**URL:** <https://forum.xbench.net/t/key-terms-understanding-memoq-tb-morphology/116>\
**Category:** General Discussion\
**Created:** [April 16, 2016, 7:08am UTC](https://forum.xbench.net/t/key-terms-understanding-memoq-tb-morphology/116 "2016-04-16T07:08:56Z")\
**Posts on this page:** 4\
**Page:** 1

<div class="post-metadata">

**Author:** ![agnus](https://avatars.discourse-cdn.com/v4/letter/a/4bbf92/32.png) [@agnus](https://forum.xbench.net/u/agnus)\
**Post date:** [April 16, 2016, 7:08am UTC](https://forum.xbench.net/t/key-terms-understanding-memoq-tb-morphology/116/1 "2016-04-16T07:08:56Z")

</div>

Hi there,

any translator to Slavic languages can tell you how useless are typical terminological QAs with their lack of understanding for any morphology. As you may or may not know, Slavic languages have a strong tendency to modify base forms of a word in a myriad ways specific to a given context, sentence, etc. In Polish one noun can have at least 14 forms. This makes a typical strict terminological QA pretty useless, as the number of false positives is overwhelming.

However, MemoQ offers a nice solution to it — I can mark basic morphology in the TB. For example I can enter there:

car — samoch|ód  
computer — komputer\*

and let it be known, that only the root „samoch” should be considered for matching, as „ód” is only present in the nominative singular, and that „komputer” can be followed by some morphological suffixes. This will remove false positives from sentences like

I didn’t have either a car or a computer.  
Nie miałem ani **samochodu** , ani **komputera**.

…which would be a case in a typical QA scenario.

And my big question is:

would it be possible for XBench to understand those two markings from MemoQ TBs and interpret them accordingly? This would be hugely beneficial mainly for Studio users, which doesn’t have any feature like this.

Best  
Jarek

[can I consider this forum as a way to make „official” feature request? I want to make sure, as this is my first post here 🙂]

---

<div class="post-metadata">

**Author:** ![pcondal](https://yyz1.discourse-cdn.com/flex035/user_avatar/forum.xbench.net/pcondal/32/305_2.png) [@pcondal](https://forum.xbench.net/u/pcondal)\
**Post date:** [April 16, 2016, 11:50am UTC](https://forum.xbench.net/t/key-terms-understanding-memoq-tb-morphology/116/2 "2016-04-16T11:50:35Z")

</div>

Posting feature requests and suggestion here is perfectly fine because we monitor this forum. Bugs or urgent support requests are a bit less useful here because they often involve sending (sensitive) files, and also because our response time is usually faster in [Support](http://www.xbench.net/index.php/support/submit-bug) than it is here.

It seems a good idea that if a user spent time configuring custom glossary entries in memoQ, we try to leverage such investment for our Key Terms QA check if the glossary in question happens to be a memoQ termbase.

We will log a task in our DB to consider implementing support for memoQ **Custom Matching** pipe (|) and asterisk (\*) characters if a memoQ TB is loaded as Key Terms.

---

<div class="post-metadata">

**Author:** ![ivaylo](https://yyz1.discourse-cdn.com/flex035/user_avatar/forum.xbench.net/ivaylo/32/14_2.png) [@ivaylo](https://forum.xbench.net/u/ivaylo)\
**Post date:** [April 9, 2018, 6:21am UTC](https://forum.xbench.net/t/key-terms-understanding-memoq-tb-morphology/116/3 "2018-04-09T06:21:04Z")

</div>

I support this request. The current Key Terms implementation gives a lot of false positives

---

<div class="post-metadata">

**Author:** ![RRLL\_LS](https://avatars.discourse-cdn.com/v4/letter/r/eb8c5e/32.png) [@RRLL\_LS](https://forum.xbench.net/u/RRLL_LS)\
**Post date:** [April 25, 2018, 10:14am UTC](https://forum.xbench.net/t/key-terms-understanding-memoq-tb-morphology/116/4 "2018-04-25T10:14:27Z")

</div>

It would be nice to have this improvement also for Multiterm glossaries. Please consider it for the next release!

Thanks in advance