Text Processing (POS, Lemma and NER)

KNIME Extensions Text Processing
#1

Hello all,

I am a new user and I try to Process my text data.
The data and text is in german language.
Currently I have a file with three columns of the type document.
The content of one of these columns I want to tag with a POS-Tagger.
I tried the Dictionary Tagger with an special model, download from the internet with a lot of pre-defined tagged words.
But the result of the node is the same of the source.
Furthermore I tried with the Bag Of Words Node to seperate all word in one special row, but with the same effect.

That is just one of my problems, but I think the biggest one.
Afterwards I want to use the Lemma of all adjectives and verbs and I want to recognize all named entities by a NER Node.

Is here anyone, who could help me?

Thank you.

#2

Hello,

Thank you for your question. Have you looked at the examples on the Examples Server and the KNIME Community Workflow Hub under 08_Other_Analytics_Types/01_Text_Processing? Here you find, for example, a workflow that tags documents with a POS-Tagger and visualizes the results in a tag cloud. The workflow is called 05_Named_Entity_Tag_Cloud. Another workflow that you find there and that could be of interest for you is 14_NER_Tagger_Model_Training. This workflow generates and evaluates an NER model.

Hope that helps,
Jeany