is there any possibility to generate a dependency tree in Knime based on the POS-Tagger results? I haven't found any included operator for that. Or is it possible to integrate an external java process, that is responsible for that processing step?
The plan is to detect references between nouns and adjectives for a sentiment analysis pipeline.
Any help is appreciated
thank you for your question. That is not possible with the existing nodes. When converting documents to a bow the order and also the sentence boundaries get lost (in the bow not in the docs themselves). The only way I see is that you implement a custom node that operates on documents and extracts relations triplets.
The Stanford NLP lib provides functionalities for triplet and relation extraction as far as I know.