I have three statements to parse:
article = ['When did you order?',\
'Did you order?',\
'Did you book the ticket?']
The result of parsetree() is wrongly tagging 'order' and 'book' as NN while it should have been VB (hence, their roles are also wrongly labeled):
WORD TAG CHUNK ROLE ID PNP LEMMA
When WRB - - - - when
did VBD VP - 1 - do
you PRP NP OBJ 1 - you
order NN NP ^ OBJ 1 - order
? . - - - - ?
---------------
WORD TAG CHUNK ROLE ID PNP LEMMA
Did VBD VP - 1 - do
you PRP NP OBJ 1 - you
order NN NP ^ OBJ 1 - order
? . - - - - ?
---------------
WORD TAG CHUNK ROLE ID PNP LEMMA
Did VBD VP - 1 - do
you PRP NP OBJ 1 - you
book NN NP ^ OBJ 1 - book
the DT NP ^ OBJ 1 - the
ticket NN NP ^ OBJ 1 - ticket
? . - - - - ?
What is the best way to correct mis-taggings like these?
Is there a grammar file that I can tweek or should I correct it in my code?
Also, does MBSP have lesser mis-tagging?