For reproducing a research paper, I have to rewrite sentences with their collocation linked by _ (exemple : this green firm uses clean energy -> this green_firm uses clean_energy).
In the paper I want to reproduce, they state that they use stanford corenlp without more explanation. Corenlp have a collocation finder class : https://nlp.stanford.edu/nlp/javadoc/javanlp-3.5.0/edu/stanford/nlp/trees/CollocationFinder.html . However, one of its arguments is WordNetConnection and I do not find any information about it. Any information on what kind of variable is WordNetConnection ? I would like to use it more or less like
CollocationFinder collocationFinder = new CollocationFinder(parseTree, wordNetConnection); collocationFinder.printCollocationStrings(pw);
I am finding so little information about it that I am wondering wether it is deprecated, but yet it was used in a 2022 published paper.
I also tried extracting the collocation with nltk but the result collocation were of low quality so I would prefer to stay with CoreNLP.