def get_top_word_frequencies(self, texts):
'''
Output: list of tuples in the format: (word,frequency,mood)
Input: texts is a list of strings.
get_top_word_frequencies finds the frequencies of words in texts.
See 'getMood' for explaination of mood.
'''
# Making the list of sentences into a list of words
temp = []
for texts_elem in texts:
temp.append(clean_tweet_text(texts_elem))
words = []
for temp_elem in temp:
for temp_elem2 in temp_elem:
words.append(temp_elem2)
# Getting the frequencies of words in a dictionary
freq = {}
for words_elem in words:
if words_elem in freq:
freq[words_elem] += 1
else:
freq[words_elem] = 1
# Getting the words sorted from highest to lowest frequency
words_sorted = sorted(freq.iteritems(),
key=operator.itemgetter(1),
reverse=True)
# Getting the final (word,frequency,mood) list
wfm = []
for words_sorted_elem in words_sorted:
wfm.append((words_sorted_elem[0],
words_sorted_elem[1],
self.wordlist.get(words_sorted_elem[0], 'None')))
wfm2 = []
for wfm_elem in wfm:
if wfm_elem[2] != 'None':
wfm2.append(wfm_elem)
return wfm2
def get_tweet_text_mood(self, text):
'''
Output: Integer, or the string 'None' if no mood is associated with it.
Input: text is a string.
get_mood gets the positivity/negativity (mood) of a string (tweet).
'''
clean_text = clean_tweet_text(text)
# The mood is the sum of the moods of each word.
text_mood = sum(map(lambda word: self.wordlist.get(word, 0), clean_text))
# Check whether any mood was found.
check = 0
for i in clean_text:
if self.wordlist.get(i, 100) != 100: # ?
check = 1
if check == 0:
text_mood = 'None'
return text_mood