def analyze_alchemyapi(self, entities):
'''find out 1.the ancestor 2.the corresponding text 3.update query'''
# the alchemyapi Query instance
query = Query(self.query.query)
query.feature_words = query.query
query.pattern = query.query
for e in entities:
ancestor = self.searchet_ontology.get_pedigree(
'http://www.alchemyapi.com/api/entity/types.html#',
e.entity_type)
text = e.entity_text
query.token_entities[text] = ancestor
query.entity_tokens[ancestor] = text
query.pattern = query.pattern.replace(text, str(ancestor))
query.feature_words = query.feature_words.replace(text, "")
# feature words
query.feature_words = [
t.strip().lower().capitalize()
for t in query.feature_words.split() if t
]
query.pattern = [p.strip() for p in query.pattern.split() if p]
return query
def analyze_freebase(self, entity):
'''docs'''
# the freebase Query instance
query = Query(self.query.query)
query.feature_words = query.query
query.pattern = query.query
ancestors = []
for t in entity.entity_type:
ancestor = self.searchet_ontology.get_pedigree(
'http://schemas.freebaseapps.com/type?id=', t)
ancestors.append(ancestor)
# dedup
ancestors = list(set(ancestors))
#special treatment for entity text
text = ''
for t in entity.entity_text:
if t in query.query.lower():
low = query.query.lower().find(t)
high = len(t) + 1
text = query.query[low:high]
break
if text:
query.token_entities[text] = ancestors
for ancestor in ancestors:
query.entity_tokens[ancestor] = text
query.feature_words = query.feature_words.replace(text, "")
query.feature_words = [
t.lower().strip().capitalize()
for t in query.feature_words.split() if t
]
if any(ancestors):
query.pattern = query.pattern.replace(text, '|'.join(ancestors))
query.pattern = [p.strip() for p in query.pattern if p]
return query
def analyze_alchemyapi(self, entities):
'''find out 1.the ancestor 2.the corresponding text 3.update query'''
# the alchemyapi Query instance
query = Query(self.query.query)
query.feature_words = query.query
query.pattern = query.query
for e in entities:
ancestor = self.searchet_ontology.get_pedigree('http://www.alchemyapi.com/api/entity/types.html#', e.entity_type)
text = e.entity_text
query.token_entities[text] = ancestor
query.entity_tokens[ancestor] = text
query.pattern = query.pattern.replace(text, str(ancestor))
query.feature_words = query.feature_words.replace(text, "")
# feature words
query.feature_words = [t.strip().lower().capitalize() for t in query.feature_words.split() if t]
query.pattern = [p.strip() for p in query.pattern.split() if p]
return query
def analyze_freebase(self, entity):
'''docs'''
# the freebase Query instance
query = Query(self.query.query)
query.feature_words = query.query
query.pattern = query.query
ancestors = []
for t in entity.entity_type:
ancestor = self.searchet_ontology.get_pedigree('http://schemas.freebaseapps.com/type?id=', t)
ancestors.append(ancestor)
# dedup
ancestors = list(set(ancestors))
#special treatment for entity text
text = ''
for t in entity.entity_text:
if t in query.query.lower():
low = query.query.lower().find(t)
high = len(t) + 1
text = query.query[low:high]
break
if text:
query.token_entities[text] = ancestors
for ancestor in ancestors:
query.entity_tokens[ancestor] = text
query.feature_words = query.feature_words.replace(text, "")
query.feature_words = [t.lower().strip().capitalize() for t in query.feature_words.split() if t]
if any(ancestors):
query.pattern = query.pattern.replace(text, '|'.join(ancestors))
query.pattern = [p.strip() for p in query.pattern if p]
return query
def __init__(self, query=None):
self.query = Query(query)
# instance of searchet-ontology-api
self.searchet_ontology = SearchetOntology(source='searchet_v2.n3',
format='n3')