def parseS(s): s = s parse = corenlp.raw_parse(s) sentences = [] for l in parse['sentences']: sentence = [] for w in l['words']: token = {} token['word'] = w[0] token['NER'] = w[1]['NamedEntityTag'] token['POS'] = w[1]['PartOfSpeech'] token['lemma'] = w[1]['Lemma'] if token['POS'] == 'FW': token['lemma']=token['word'] if token['word'] == '-LRB-': token['word'] = '(' if token['word'] == '-RRB-': token['word'] = ')' token['WS'] = simp.getSynset(toString2(l['words']), token['word']) sentence.append(token) sentences.append(sentence) if 'coref' in parse: coref(parse['coref'],sentences) return sentences
def getSynset(s,w): return simp.getSynset(s,w) '''wpos = ""