def parse_add(self, url, response_body):
a=newspaper.Article("")
a.is_downloaded = True
a.html = response_body
a.parse()
a.nlp()
img_link = a.top_img
named, persons, orgs= self.get_named_entities(a.text)
print "alma base search started"
persons = [person for person in persons if alma.search(person) > 0]
orgs = [org for org in orgs if alma.search(org) > 0]
print "alma base search ended"
author="default"
try:
author=a.author[0]
except:
print "Not found"
art=self.add_article(a.title,a.summary,url,author,img_link, named, persons, orgs)
# test_keywords(art,newsClassifier)
return art
def train(url,keywords):
# for url in urls:
a=newspaper.Article(url)
a.download()
a.parse()
a.nlp()
img_link = a.top_img
named, persons, orgs= get_named_entities(a.text)
print "alma base search started"
persons = [person for person in persons if alma.search(person) > 0]
orgs = [org for org in orgs if alma.search(org) > 0]
print "alma base search stoped"
author="default"
try:
author=a.author[0]
except:
print "Not found"
art=add_article(a.title,a.summary,url,author,img_link, named, persons, orgs)
# test_keywords(art,newsClassifier)
return keywords
def train(url,keywords):
# for url in urls:
a=newspaper.Article(url)
a.download()
a.parse()
a.nlp()
img_link = a.top_img
named, persons, orgs= get_named_entities(a.text)
print "alma base search started"
persons = [person for person in persons if alma.search(person) > 0]
orgs = [org for org in orgs if alma.search(org) > 0]
print "alma base search ended"
author="default"
try:
author=a.author[0]
except:
print "Not found"
art=add_article(a.title,a.summary,url,author,img_link, named, persons, orgs)
# test_keywords(art,newsClassifier)
return keywords
