0

I am parsing Noun Chunks with the help of Spacy and I want to "normalize" those Noun Chunks, that means I want to change a given Noun from its grammatikal case to another grammatikal case (f.e. Genitiv to Nominativ). All I managed to achive so far ist to label the noun chunks using the Spacy morphologizer. Since this problem is quite unique to the german language i was not able to find much about this problem on the internet.

An Example would be turning this:

"dortiger Verfügung" dortiger Case=Dat|Degree=Pos|Gender=Fem|Number=Sing Verfügung Case=Dat|Gender=Fem|Number=Sing

into this: "dortige Verfügung" dortige Case=Nom|Degree=Pos|Gender=Fem|Number=Sing Verfügung Case=Nom|Gender=Fem|Number=Sing

schieslu
  • 1
  • I dont understand what you mean, I think my problem is quite specific. I need to change the morphology of a given word with a given set of rules, preferably using the Spacy morphologizer in a reverse way but I am open to any other suggestions that lead to the same result. – schieslu Nov 28 '22 at 11:00
  • How about adding the precise program you used to label the nounchunks, and the actual input you provided. – rici Nov 29 '22 at 02:24
  • As mentioned in spaCy discussions, transforming a noun into a given case is called "surface realization". However, turning a noun into a default case (such as nominative) is usually covered by lemmatization. https://github.com/explosion/spaCy/discussions/11883 – polm23 Nov 29 '22 at 03:32
  • While the exact words and cases of German language are unique, the process is not specific to German. Basically the same question here. https://stackoverflow.com/questions/45590278/how-to-inverse-lemmatization-process-given-a-lemma-and-a-token/45591295#45591295 – polm23 Nov 29 '22 at 03:34

0 Answers0