def on_status(self, status):
try:
if not status.retweeted and not status.in_reply_to_status_id and not status.in_reply_to_user_id and not hasattr(status, 'retweeted_status'):
features = tweet_features.get_tweet_features(status.text, worstfeaturesfilter)
print features
mood = classifier.classify(features)
if mood == 'negative':
if len(jokes) > 0:
#TODO: respect word boundaries
jokechunks = jokes.pop()
pickle.dump(jokes,open('jokes.pickle', 'w'))
for chunk in jokechunks:
api.update_status(chunk)
else:
print "we're out of jokes!!"
print self.status_wrapper.fill(status.text)
print '\n %s %s via %s\n' % (status.author.screen_name, status.created_at, status.source)
print "mood classified as: %s" % mood
classifier.explain(features)
except Exception, e:
# Catch any unicode errors while printing to console
# and just ignore them to avoid breaking application.
print e
pass
tweets.append((tweet.text, label))
elif label == "negative":
tweets.append((tweet.text, label))
labelcount[label] += 1
random.shuffle(tweets)
print labelcount
# split in to training and test sets
random.shuffle(tweets)
num_train = int(0.8 * len(tweets))
# fvecs = [(tweet_features.make_tweet_dict(t),s) for (t,s) in tweets]
fvecs = [(tweet_features.get_tweet_features(t, set()), s) for (t, s) in tweets]
v_train = fvecs[0:num_train]
# v_train = fvecs
v_test = fvecs[num_train : len(tweets)]
# extract best word features
word_fd = FreqDist()
label_word_fd = ConditionalFreqDist()
for (feats, label) in fvecs:
for key in feats:
if feats[key]:
word_fd.inc(key)
label_word_fd[label].inc(key)
if label == 'positive':
tweets.append((tweet.text, label))
elif label == 'negative':
tweets.append((tweet.text, label))
labelcount[label] += 1
random.shuffle(tweets)
print labelcount
# split in to training and test sets
random.shuffle(tweets)
num_train = int(0.8 * len(tweets))
#fvecs = [(tweet_features.make_tweet_dict(t),s) for (t,s) in tweets]
fvecs = [(tweet_features.get_tweet_features(t, set()), s) for (t, s) in tweets]
v_train = fvecs[0:num_train]
#v_train = fvecs
v_test = fvecs[num_train:len(tweets)]
#extract best word features
word_fd = FreqDist()
label_word_fd = ConditionalFreqDist()
for (feats, label) in fvecs:
for key in feats:
if feats[key]:
word_fd.inc(key)
label_word_fd[label].inc(key)
pos_word_count = label_word_fd['positive'].N()