def small_text():
sentence1 = "Travel kills time."
sentence2 = "France is a nice country."
text = sentence1 + " " + sentence2
term_extractor = C_NC_TermExtractor(text)
terms = term_extractor.compute_cnc()
former = ConceptFormer()
former.form_concepts(terms)
tripels = list(RelationExtractor.find_realation(text))
former.find_hearst_concepts(tripels)
print "Taxonomy: "
pprint(former.get_taxonomy())
concepts, relations = [], []
for concept in list(former.get_taxonomy()):
concepts.append(" ".join(concept.name))
relations += concept.make_tripels()
print "no con.: " + str(len(concepts))
print "no rel.: " + str(len(relations))
utils.dot_to_image(utils.taxonomy_to_dot(concepts, relations), 'france')
def small_text():
sentence1 = "Travel kills time."
sentence2 = "France is a nice country."
text = sentence1 + " " + sentence2
term_extractor = C_NC_TermExtractor(text)
terms = term_extractor.compute_cnc()
former = ConceptFormer()
former.form_concepts(terms)
tripels = list(RelationExtractor.find_realation(text))
former.find_hearst_concepts(tripels)
print "Taxonomy: "
pprint(former.get_taxonomy())
concepts, relations = [], []
for concept in list(former.get_taxonomy()):
concepts.append(" ".join(concept.name))
relations += concept.make_tripels()
print "no con.: " + str(len(concepts))
print "no rel.: " + str(len(relations))
utils.dot_to_image(utils.taxonomy_to_dot(concepts, relations), 'france')
import concept_former as cf import term import hearst_patterns as hp former = cf.conceptFormer() text = 'The monkey bites the snake.' pattern = hp.find_realation(text) result = former.find_hearst_concepts(pattern) print result for concept in list(result): print str(concept) + ' ' + str(concept.get_relations())
import utils import preprocessor as pp from tree_combinations import numerate_non_terminals import hearst_patterns file_name = 'LeonHitsKai' text = 'Leon hits Kai.' print 'get relations by applying hearst patterns' relations = hearst_patterns.find_realation(text) print relations print print 'generate dot code' dot_code = utils.list_of_tripels_to_dot(relations) print dot_code print print 'convert dot code to image' utils.dot_to_image(dot_code, file_name + '_relations')
import utils import preprocessor as pp from tree_combinations import numerate_non_terminals import hearst_patterns file_name = 'LeonHitsKai' text = 'Leon hits Kai.' print 'get relations by applying hearst patterns' relations = hearst_patterns.find_realation(text) print relations print print 'generate dot code' dot_code = utils.list_of_tripels_to_dot(relations) print dot_code print print 'convert dot code to image' utils.dot_to_image(dot_code, file_name + '_relations')
import corpus
import hearst_patterns as hp
import utils
file_name = 'simple'
#text = corpus.CorpusReader().get_corpus()
text = 'Leon hits Kai. Marry f***s John. Kai greets Marry. John greets Leon.'
relations = hp.find_realation(text)
print
for r in relations:
print r
if utils.which('dot'):
dot_code = utils.list_of_tripels_to_dot(relations)
utils.dot_to_image(dot_code, file_name + '_relations')
else:
print "didn't find dot"
import concept_former as cf import term import hearst_patterns as hp former = cf.conceptFormer() text = 'The monkey bites the snake.' pattern = hp.find_realation(text) result = former.find_hearst_concepts(pattern) print result for concept in list(result): print str(concept) + ' ' + str(concept.get_relations())