Besetzung, Wirkung

[GoogleCodeExporter]

If a linguistic problem:
What wordform makes the faulty analysis occur?

>Wirkung
Wirkung<+NN><Fem><Gen><Sg>
Wirkung<+NN><Fem><Akk><Sg>
Wirkung<+NN><Fem><Dat><Sg>
Wirkung<+NN><Fem><Nom><Sg>
wirken<V>ung<SUFF><+NN><Fem><Gen><Sg>
wirken<V>ung<SUFF><+NN><Fem><Akk><Sg>
wirken<V>ung<SUFF><+NN><Fem><Dat><Sg>
wirken<V>ung<SUFF><+NN><Fem><Nom><Sg>

>Besetzung
be<PREF>setzen<V>ung<SUFF><+NN><Fem><Gen><Sg>
be<PREF>setzen<V>ung<SUFF><+NN><Fem><Akk><Sg>
be<PREF>setzen<V>ung<SUFF><+NN><Fem><Dat><Sg>
be<PREF>setzen<V>ung<SUFF><+NN><Fem><Nom><Sg>
besetzen<V>ung<SUFF><+NN><Fem><Gen><Sg>
besetzen<V>ung<SUFF><+NN><Fem><Akk><Sg>
besetzen<V>ung<SUFF><+NN><Fem><Dat><Sg>
besetzen<V>ung<SUFF><+NN><Fem><Nom><Sg>

In my opinion Besetzung is a usual noun, therefore I would expect, that 
morphist, similarly to Wirkung also lists the noun form without suffix of ung 
an other complications.

In fact, what a user needs, if he gets Besetzungen and consults morphisto, 
morphisto passes back the basic form Besetzung; this is not the case at 
present. 

Original issue reported on code.google.com by eleonor...@gmx.net on 19 Aug 2011 at 3:45

[GoogleCodeExporter]

Solution: add
<Base_Stems>Besetzung<NN><base><nativ><NFem_0_en>
to lexicon

Original comment by eleonor...@gmx.net on 19 Aug 2011 at 9:29

[GoogleCodeExporter]

I disagree with that view. "Besetzung" is clearly a derivate of "besetzen" if 
not even of "setzen". For me, it is a great advantage that morphisto has not a 
similarly blown-up lexicon like other finite-state morphologys. For base form 
reduction you can use "fst-infl2 -b" and easily get rid of the infinitive 
suffix.

Original comment by wuerz...@gmail.com on 24 Aug 2011 at 4:40

[GoogleCodeExporter]

Original comment by wuerz...@gmail.com on 25 Aug 2011 at 9:52