Open jbmartin opened 10 years ago
Thanks for the suggestion and the link! It is a nice method. I don't quite see how to put it into an automatic, algorithmic form to add to Expresso. Passive voice and weak verbs are already highlighted. I guess, I can highlight prepositions as well. If there is a cluster of them nearby, the sentence can probably be improved. Does this sound right to you?
I agree: designing an algorithm that performs the paramedic method would be a little tricky. Possible metrics could be ratio of prepositions to words in a sentence and does a sentence start with a dependent clause. Does the NLTK package have a way to identify dependent clauses?
It looks like NLTK does not have a good dependency parser of sentences but Stanford does. I am thinking of adding it to Expresso in April-May, when I finish my PhD.
Good luck on your defense! If I get a little bit of time, I can also try to take a stab at it.
Thanks! Yes, feel free to add functionality. I appreciate other people's input.
Might be useful to implement the paramedic method, which could highlight prepositions (independent of stop words) and sentences starting with dependent clauses.