def test_tokenizer(self):
# Assert that french sentiment() uses French tokenizer. ("t'aime" => "t' aime").
v1 = fr.sentiment("je t'aime")
v2 = fr.sentiment("je ne t'aime pas")
self.assertTrue(v1[0] > 0)
self.assertTrue(v2[0] < 0)
self.assertTrue(v1.assessments[0][0] == ["aime"])
self.assertTrue(v2.assessments[0][0] == ["ne", "aime"])
def test_tokenizer(self):
# Assert that french sentiment() uses French tokenizer. ("t'aime" => "t' aime").
v1 = fr.sentiment("je t'aime")
v2 = fr.sentiment("je ne t'aime pas")
self.assertTrue(v1[0] > 0)
self.assertTrue(v2[0] < 0)
self.assertTrue(v1.assessments[0][0] == ["aime"])
self.assertTrue(v2.assessments[0][0] == ["ne", "aime"])
def emotion_ts_from_text(tier, nlp):
x = []
y = []
z = []
for sppasOb in tier:
label, [start, stop], [start_r, stop_r] = get_interval(sppasOb)
i = (start + stop) / 2.0
if label in ["#", "", " ", "***", "*"]:
polarity = 0
subjectivity = 0
else:
polarity_and_sunjectivity = sentiment(label)
polarity = polarity_and_sunjectivity[0]
subjectivity = polarity_and_sunjectivity[1]
while (i < stop):
x.append(i)
y.append(polarity)
z.append(subjectivity)
i = i + ((start + stop) / 10.0)
return x, y, z
def test_sentiment(self):
# Assert < 0 for negative adjectives and > 0 for positive adjectives.
self.assertTrue(fr.sentiment("fabuleux")[0] > 0)
self.assertTrue(fr.sentiment("terrible")[0] < 0)
# Assert the accuracy of the sentiment analysis.
# Given are the scores for 1,500 book reviews.
# The baseline should increase (not decrease) when the algorithm is modified.
from pattern.db import Datasheet
from pattern.metrics import test
reviews = []
for review, score in Datasheet.load(os.path.join(PATH, "corpora", "polarity-fr-amazon.csv")):
reviews.append((review, int(score) > 0))
A, P, R, F = test(lambda review: fr.positive(review), reviews)
self.assertTrue(A > 0.75)
self.assertTrue(P > 0.76)
self.assertTrue(R > 0.73)
self.assertTrue(F > 0.75)
print "pattern.fr.sentiment()"
def test_sentiment(self):
# Assert < 0 for negative adjectives and > 0 for positive adjectives.
self.assertTrue(fr.sentiment("fabuleux")[0] > 0)
self.assertTrue(fr.sentiment("terrible")[0] < 0)
# Assert the accuracy of the sentiment analysis.
# Given are the scores for 1,500 book reviews.
# The baseline should increase (not decrease) when the algorithm is modified.
from pattern.db import Datasheet
from pattern.metrics import test
reviews = []
for review, score in Datasheet.load(os.path.join(PATH, "corpora", "polarity-fr-amazon.csv")):
reviews.append((review, int(score) > 0))
A, P, R, F = test(lambda review: fr.positive(review), reviews)
self.assertTrue(A > 0.75)
self.assertTrue(P > 0.76)
self.assertTrue(R > 0.73)
self.assertTrue(F > 0.75)
print "pattern.fr.sentiment()"
if i == 0:
header = row
else:
for c in string.punctuation:
body = body.replace(c,' ')
for c in string.whitespace:
body = body.replace(c,' ')
for c in ["'"]:
body = body.replace(c,' ')
words = clean_body([n.lower() for n in body.split(" ")])
comment = " ".join(words)
polarity, subjectivity = sentiment(comment)
vote = (polarity+subjectivity)
subjectivity = subjectivity+1.0000001
polarity = polarity*subjectivity
try:
mood[polarity].append(row)
except:
mood[polarity] = [row]
#.append(comment)
#print polarity,"\t\t", comment[:50],"\t\t", len(comment[:50]),"\t\t"
#print stem(s, stemmer=PORTER)
words.extend(words)
i =+ 1
import pandas as pd
import sys
store = 'tweets.csv'
df = pd.read_csv(store,
sep=';',
names=['brand', 'id', 'username', 'date', 'tweet', 'source'])
brand1 = df[df.brand == 'SNCB']
brand2 = df[df.brand == 'SNCF']
from pattern.fr import parse, sentiment, ngrams, pprint
from pattern.web import URL, plaintext
from pattern.metrics import readability
b1_read = 0.0
b1_sent = 0.0
for tweet in brand1.tweet:
b1_sent += sentiment(plaintext(tweet))[0]
b1_read += readability(tweet)
b2_read = 0.0
b2_sent = 0.0
for tweet in brand2.tweet:
b2_sent += sentiment(plaintext(tweet))[0]
b2_read += readability(tweet)
print 'SNCB: %f' % b1_sent
print 'SNCF: %f' % b2_sent
