def update_bigram_hash(self,word_array,bigrams_hash):
bigrams=utils.generate_bigrams(word_array)
for key in bigrams:
if key in bigrams_hash.keys():
bigrams_hash[key]+=1
else:
bigrams_hash[key]={}
bigrams_hash[key]=1
def identify_interaction(self,path_to_test_file):
document_data=open(path_to_test_file,'rb')
xml_data=parse(document_data)
sentences = xml_data.getElementsByTagName("sentence")
result={}
for sentence in sentences:
entity_collection={}
sentence_attrs = dict(sentence.attributes.items())
sentence_text=sentence_attrs["text"]
sentence_id=sentence_attrs["id"]
entities = sentence.getElementsByTagName("entity")
for entity in entities:
entity_attrs = dict(entity.attributes.items())
id=entity_attrs["id"]
text=(entity_attrs["text"]).lower()
type=(entity_attrs["type"]).lower()
entity_collection[id]={}
entity_collection[id]={"text":text,"type":type}
words=utils.tokenize_string_without_punctuations(sentence_text)
bigrams=utils.generate_bigrams(words)
trigrams=utils.generate_trigrams(words)
result_found=0
for bigram in bigrams:
if bigram in self.bigrams_based_on_score.keys():
result[sentence_id]={}
result[sentence_id]=self.bigrams_based_on_score[bigram]
result_found=1
print "bigram as a result of which classification is done:"+bigram
break
if result_found!=1:
result[sentence_id]="false"
#pprint.pprint(self.bigrams_based_on_score)
document_data.close()
return result
