def build_dissimilarity_graph_using_wordnet(self, tokenised_sentence_dict): sentence_keys = tokenised_sentence_dict.keys() synonyms = Synonyms(tokenised_sentence_dict.values()) self.dissimilar_sentences.add_nodes(sentence_keys) connections = ConnectedNodes() edges = list() for every_key, tokens in tokenised_sentence_dict.iteritems(): for other_key, other_tokens in tokenised_sentence_dict.iteritems(): if connections.not_connected(every_key, other_key): score = synonyms.dissimilarity_score(tokens, other_tokens) if score > self.DISSIMILARITY_THRESHOLD: edge = (every_key, other_key, {"weight": score}) edges.append(edge) connections.add((every_key, other_key)) self.dissimilar_sentences.add_edges(edges)
def build_dissimilarity_graph(self, tokenised_sentence_dict): print "Building Dissimilar Sentences Graph" sentence_keys = tokenised_sentence_dict.keys() transformed_sentences = self.transformation(tokenised_sentence_dict) self.dissimilar_sentences.add_nodes(sentence_keys) connections = ConnectedNodes() edges = list() for every_key, tokens in tokenised_sentence_dict.iteritems(): for other_key, other_tokens in tokenised_sentence_dict.iteritems(): if connections.not_connected(every_key, other_key): score = transformed_sentences.dissimilarity_score(tokens, other_tokens) if score > self.DISSIMILARITY_THRESHOLD: edge = (every_key, other_key, {"weight": score}) edges.append(edge) connections.add((every_key, other_key)) self.dissimilar_sentences.add_edges(edges) print "Dissimilar Sentences Graph build complete"