Accessing Stanford Core NLP Coreference Chain output in Ruby -

i trying use stanford core nlp :coref_chain list of entity mentions out of text. when run code:

text = 'angela merkel met nicolas sarkozy on january 25th in ' +    'berlin discuss new austerity package. sarkozy ' +    'looked pleased, merkel dismayed.'  pipeline =  stanfordcorenlp.load(:tokenize, :ssplit, :pos, :lemma, :parse, :ner, :dcoref) text = stanfordcorenlp::annotation.new(text) pipeline.annotate(text)    puts text.get(:coref_chain) 

i output:

{1=chain1-["angela merkel" in sentence 1, "merkel" in sentence 2], 3=chain3-["january 25th" in sentence 1], 4=chain4-["berlin" in sentence 1], 5=chain5-["nicolas sarkozy on january 25th" in sentence 1, "sarkozy" in sentence 2], 6=chain6-["a new austerity package" in sentence 1]} 

is hash? according documentation on stanford site, should able access these values through attribute names no combination has worked me. in fact, adding other to_s yields "no method found" error.

does 1 know how names out of this? "angela merkel" example? in best case scenario, start, end , heads well.

look @ the example code using gem. have loop on returned value of text.get:

text.get(:sentences).each |sentence| 

you can loop on sentences returned , extract further information.

what think hash isn't hash. it's string representation of internal structure they're using created when told puts, similar inspect return.