how to get parse tree using python nltk?

9
votes

Given the following sentence:

The old oak tree from India fell down.

How can I get the following parse tree representation of the sentence using python NLTK?

(ROOT (S (NP (NP (DT The) (JJ old) (NN oak) (NN tree)) (PP (IN from) (NP (NNP India)))) (VP (VBD fell) (PRT (RP down)))))

I need a complete example which I couldn't find in web!

Edit

I have gone through this book chapter to learn about parsing using NLTK but the problem is, I need a grammar to parse sentences or phrases which I do not have. I have found this stackoverflow post which also asked about grammar for parsing but there is no convincing answer there.

So, I am looking for a complete answer that can give me the parse tree given a sentence.

2
pythonnltk
NLTK has several parsers (nltk.org/api/nltk.parse.html) IIRC I for preference use the Stanford one. - Lyndon White
can you give me an example code? i am really having tough time in cracking this. - Wasi Ahmad
I've got to run off, if no one post anything in the next 12 hours I'll come back and post something. It has been a while I'ld have to dig up some of my old code (and translate it from julia to python probably). - Lyndon White
@WasiAdmad, if you're actually "having a tough time", show your code so far and ask a question about the problem you encounter. - alexis

2 Answers

8
votes

Here is alternative solution using StanfordCoreNLP instead of nltk. There are few library that build on top of StanfordCoreNLP, I personally use pycorenlp to parse the sentence.

First you have to download stanford-corenlp-full folder where you have *.jar file inside. And run the server inside the folder (default port is 9000).

export CLASSPATH="`find . -name '*.jar'`"
java -mx4g -cp "*" edu.stanford.nlp.pipeline.StanfordCoreNLPServer [port?] # run server

Then in Python, you can run the following in order to tag the sentence.

from pycorenlp import StanfordCoreNLP
nlp = StanfordCoreNLP('http://localhost:9000')

text = "The old oak tree from India fell down."

output = nlp.annotate(text, properties={
  'annotators': 'parse',
  'outputFormat': 'json'
})

print(output['sentences'][0]['parse']) # tagged output sentence
1
votes

Older question, but you can use nltk together with the bllipparser. Here is a longer example from nltk. After some fiddling I myself used the following:

To install (with nltk already installed):

sudo python3 -m nltk.downloader bllip_wsj_no_aux
pip3 install bllipparser

To use:

from nltk.data import find
from bllipparser import RerankingParser

model_dir = find('models/bllip_wsj_no_aux').path
parser = RerankingParser.from_unified_model_dir(model_dir)

best = parser.parse("The old oak tree from India fell down.")

print(best.get_reranker_best())
print(best.get_parser_best())

Output:

-80.435259246021 -23.831876011253 (S1 (S (NP (NP (DT The) (JJ old) (NN oak) (NN tree)) (PP (IN from) (NP (NNP India)))) (VP (VBD fell) (PRT (RP down))) (. .)))
-79.703612178593 -24.505514522222 (S1 (S (NP (NP (DT The) (JJ old) (NN oak) (NN tree)) (PP (IN from) (NP (NNP India)))) (VP (VBD fell) (ADVP (RB down))) (. .)))