def get_language_results(tweet_text_array):
parsed_languages = []
raw_results = indicoio.language(tweet_text_array, top_n=1)
for result in raw_results:
field, value = result.items()[0]
parsed_languages.append(field)
return parsed_languages
def analysis(data):
sentiment = ind.sentiment_hq(data)
tags = sort(ind.text_tags(data))
languages = sort(ind.language(data))
politics = sort(ind.political(data))
keywords = sort(ind.keywords(data))
names = sort(ind.named_entities(data))
print "Sentiment", sentiment
print "\n\n\nTags"
for t in tags:
print t[0], float(t[1]) * 100
print "\n\n\nLanguages"
for l in languages:
print l[0], float(l[1]) * 100
print "\n\n\nPolitical"
for p in politics:
print p[0], float(p[1]) * 100
print "\n\nkeywords"
for k in keywords:
print k[0], float(k[1]) * 100
def convertData(x):
data = x
name = data['name']
lang = indicoio.language(data['words'])
if lang['English'] > lang['Spanish']:
language = 'english'
if lang['English'] < lang['Spanish']:
language = 'spanish'
sent = round(indicoio.sentiment(data['words']), 2)
words = data['words'].split()
numwords = len(words)
totalL = 0
for j in words:
totalL += len(j)
meanLW = round((totalL / numwords), 2)
#Encoding Total-lenght
# Values :
# 0 : short < 20
# 1 : medium > 20
# 2 : long > 40
if totalL < 20:
totalL = 0
if totalL > 40:
totalL = 2
if totalL > 20:
totalL = 1
#Encoding numwords
if numwords < 5:
numwords = 0
if numwords > 10:
numwords = 2
if numwords >= 5:
numwords = 1
#Encoding sentiment
#Values :
# 1 : Possitive
# 2 : Negative
# 0 : Neutral
if sent > 0.6:
sent = 1
if sent < 0.4:
sent = 2
if sent < 1:
sent = 0
if language == 'english':
language = 1
if language == 'spanish':
language = 0
sent = 0
cad = [name, language, totalL, meanLW, sent, numwords]
return cad
def isNotEnglish(text):
language = indicoio.language(text)
print(language)
testy = sorted(language.keys(), key=lambda x: language[x], reverse=True)[:1] < 0.5
print(testy)
print("!!")
return testy
def test_language(self):
language_set = set([
'English', 'Spanish', 'Tagalog', 'Esperanto', 'French', 'Chinese',
'French', 'Bulgarian', 'Latin', 'Slovak', 'Hebrew', 'Russian',
'German', 'Japanese', 'Korean', 'Portuguese', 'Italian', 'Polish',
'Turkish', 'Dutch', 'Arabic', 'Persian (Farsi)', 'Czech',
'Swedish', 'Indonesian', 'Vietnamese', 'Romanian', 'Greek',
'Danish', 'Hungarian', 'Thai', 'Finnish', 'Norwegian', 'Lithuanian'
])
language_dict = language('clearly an english sentence')
self.assertEqual(language_set, set(language_dict.keys()))
assert language_dict['English'] > 0.25
def test_language(self):
language_set = set([
'English',
'Spanish',
'Tagalog',
'Esperanto',
'French',
'Chinese',
'French',
'Bulgarian',
'Latin',
'Slovak',
'Hebrew',
'Russian',
'German',
'Japanese',
'Korean',
'Portuguese',
'Italian',
'Polish',
'Turkish',
'Dutch',
'Arabic',
'Persian (Farsi)',
'Czech',
'Swedish',
'Indonesian',
'Vietnamese',
'Romanian',
'Greek',
'Danish',
'Hungarian',
'Thai',
'Finnish',
'Norwegian',
'Lithuanian'
])
language_dict = language('clearly an english sentence')
self.assertEqual(language_set, set(language_dict.keys()))
assert language_dict['English'] > 0.25
def test_batch_language(self):
test_data = ['clearly an english sentence']
response = language(test_data, api_key=self.api_key)
self.assertTrue(isinstance(response, list))
self.assertTrue(response[0]['English'] > 0.25)
def test_language(self):
language_dict = language('clearly an english sentence')
self.assertEqual(LANGUAGES, set(language_dict.keys()))
assert language_dict['English'] > 0.25
def test_batch_language(self):
test_data = ["clearly an english sentence"]
response = language(test_data)
self.assertTrue(isinstance(response, list))
self.assertTrue(response[0]["English"] > 0.25)
def get_language():
if request.method == 'POST':
data = dict(request.form)['data_to_analyze']
return json.dumps({
'language': sort(indicoio.language(data)[0])[0]
})
def test_language(self):
language_dict = language('clearly an english sentence')
self.assertEqual(LANGUAGES, set(language_dict.keys()))
assert language_dict['English'] > 0.25
