def test_feature():
microblogId, microblog_text, polarity = "234", "我不喜欢你， 你这个愚昧的家伙，脑子有毛病", 1
feature_extracter = FeatureExtractor()
words, taggings = feature_extracter.pos_tagging(microblog_text)
posCount, negCount = feature_extracter.polarity_count(microblog_text)
single_microblog = Microblog(microblogId=microblogId, text=microblog_text, polarity=polarity, negCount=negCount,
posCount=posCount, words=words, taggings=taggings, microblogType=1, topic=None, sentiment=None)
single_microblog.save()
def test_feature():
microblogId, microblog_text, polarity = "234", "我不喜欢你， 你这个愚昧的家伙，脑子有毛病", 1
feature_extracter = FeatureExtractor()
words, taggings = feature_extracter.pos_tagging(microblog_text)
posCount, negCount = feature_extracter.polarity_count(microblog_text)
single_microblog = Microblog(microblogId=microblogId,
text=microblog_text,
polarity=polarity,
negCount=negCount,
posCount=posCount,
words=words,
taggings=taggings,
microblogType=1,
topic=None,
sentiment=None)
single_microblog.save()
def complete_microblog():
microblogs = Microblog.objects(polarity='pos', microblogType='training')
print(len(microblogs))
new_microblogs = []
for microblog in microblogs:
newone = Microblog(microblogId=microblog.microblogId,
text=microblog.text,
polarity=microblog.polarity,
negCount=microblog.negCount,
posCount=microblog.posCount,
words=microblog.words,
taggings=microblog.taggings,
microblogType=microblog.microblogType,
topic='',
sentiment='')
new_microblogs.append(newone)
Microblog.objects.insert(new_microblogs)
def complete_microblog():
microblogs = Microblog.objects(polarity='pos', microblogType='training')
print(len(microblogs))
new_microblogs = []
for microblog in microblogs:
newone = Microblog(microblogId=microblog.microblogId, text=microblog.text, polarity=microblog.polarity, negCount=microblog.negCount,
posCount=microblog.posCount, words=microblog.words, taggings=microblog.taggings, microblogType=microblog.microblogType, topic='', sentiment='')
new_microblogs.append(newone)
Microblog.objects.insert(new_microblogs)
def get_feature_set(microblogType):
microblogs = Microblog.objects()
words_features = get_words_features_pickle()
feature_sets = [(feature_filter(microblog.words, words_features), microblog.polarity) for microblog in microblogs]
return feature_sets
def classify_testing():
microblogs = Microblog.objects(microblogType='testing')
baseline_method(microblogs)
latent_polarity_method(microblogs)
test_set = get_feature_set('testing')
for (name, input_path) in classifier_path_list:
with open(input_path, 'rb') as input_classifier:
classifier = pickle.load(input_classifier)
save_testing_result(classifier, test_set, name)
def new_microblog():
form = MicroblogForm()
if form.validate_on_submit():
microblog = Microblog(body=form.microblog.data,author=current_user)
db.session.add(microblog)
db.session.commit()
flash('Your microblog is now live!')
return redirect(url_for('main.new_microblog'))
page = request.args.get('page',1,type=int)
pagination = current_user.followed_microblogs().paginate(
page,current_app.config['POSTS_PER_PAGE'])
microblogs = pagination.items
return render_template('microblog/new_microblog.html',form=form, microblogs=microblogs, pagination=pagination)
def pickle_words_features(microblogType):
microblogs = Microblog.objects()
#pickle pattern
pattern_induct(microblogs)
all_words = []
for microblog in microblogs:
all_words.extend(microblog.words)
all_words = nltk.FreqDist(all_words)
words_features = list(all_words.keys())[:3000]
with open(WORDS_FEATURES_PATH, 'wb') as output_file:
pickle.dump(words_features, output_file)
def microblog_data_handler(microblog_type):
data_filter = DataFilter()
microblogs = data_filter.read_and_filter_api_microblog_data(microblog_type)
result = []
feature_extractor = FeatureExtractor()
for polarity in microblogs:
microblog_list = microblogs[polarity]
for microblog in microblog_list:
#feature extractor
microblogId, microblog_raw_text, microblog_text, polarity = microblog[
0], microblog[1], microblog[2], microblog[3]
posCount, negCount = feature_extractor.polarity_count(
microblog_text)
words, taggings = feature_extractor.pos_tagging(microblog_text)
raw_words, raw_taggings = feature_extractor.pos_tagging(
microblog_raw_text)
single_microblog = Microblog(microblogId=microblogId,
text=microblog_text,
polarity=polarity,
microblogType=microblog_type,
topic='',
posCount=posCount,
negCount=negCount,
words=words,
taggings=taggings,
raw_words=raw_words,
raw_taggings=raw_taggings,
sentiment='')
result.append(single_microblog)
return result
def delete_microblog_by_id(microblog_id):
microblog = Microblog.objects(microblogId=microblog_id).first()
if microblog:
microblog.delete()
def delete_all_microblogs_by_type(microblog_type):
microblogs = Microblog.objects(microblogType=microblog_type)
if microblogs:
microblogs.delete()
def delete_all_microblogs_by_type(microblog_type):
microblogs = Microblog.objects(microblogType=microblog_type)
if microblogs:
microblogs.delete()
def delete_microblog_by_id(microblog_id):
microblog = Microblog.objects(microblogId=microblog_id).first()
if microblog:
microblog.delete()