0.900169

0.162830

0.335000

0.692300

1.000000

1.000000

1.000000

0.638298

0.330440

0.000000

0.360600

0.670600

1.000000

0

0

0

0

0

0

1.000000

44

https://colab.research.google.com/drive/1SBn5i_qSYKVq0ThDZh0ZG2WC2-C7czEx#printMode=true

1/7

13/11/2021 10:13

TF-IDF for Sentiment Analysis.ipynb - Colaboratory

1 airline_tweets.info()

<class 'pandas.core.frame.DataFrame'>

RangeIndex: 14640 entries, 0 to 14639

Data columns (total 15 columns):

Column Non-Null Count Dtype

--- ------ -------------- -----

0 tweet_id 14640 non-null int64

1 airline_sentiment 14640 non-null object

2 airline_sentiment_confidence 14640 non-null float64

3 negativereason 9178 non-null object

4 negativereason_confidence 10522 non-null float64

5 airline 14640 non-null object

6 airline_sentiment_gold 40 non-null object

7 name 14640 non-null object

8 negativereason_gold 32 non-null object

9 retweet_count 14640 non-null int64

10 text 14640 non-null object

11 tweet_coord 1019 non-null object

12 tweet_created 14640 non-null object

13 tweet_location 9907 non-null object

14 user_timezone 9820 non-null object

dtypes: float64(2), int64(2), object(11)

memory usage: 1.7+ MB

1

2

3

4

5

6

plot_size = plt.rcParams["figure.figsize"]

print(plot_size[0])

print(plot_size[1])

plot_size[0] = 14

plot_size[1] = 8

plt.rcParams["figure.figsize"] = plot_size

6.0

4.0

1 airline_tweets['airline'].unique()

array(['Virgin America', 'United', 'Southwest', 'Delta', 'US Airways',

'American'], dtype=object)

1 count=airline_tweets.airline.value_counts()

1 print(count)

United 3822

US Airways 2913

American 2759

Southwest 2420

Delta 2222

Virgin America 504

Name: airline, dtype: int64

1 airline_tweets.airline.value_counts().plot(kind='pie', autopct='%1.0f%%')

https://colab.research.google.com/drive/1SBn5i_qSYKVq0ThDZh0ZG2WC2-C7czEx#printMode=true

2/7

13/11/2021 10:13

TF-IDF for Sentiment Analysis.ipynb - Colaboratory

2 # Cette figure représente la répartition des données par compagines aériennes

<matplotlib.axes._subplots.AxesSubplot at 0x7f7545b94790>

1 airline_tweets.airline_sentiment.value_counts().plot(kind='pie', autopct='%1.0f%%', col

<matplotlib.axes._subplots.AxesSubplot at 0x7f75448115d0>

https://colab.research.google.com/drive/1SBn5i_qSYKVq0ThDZh0ZG2WC2-C7czEx#printMode=true

3/7

13/11/2021 10:13

TF-IDF for Sentiment Analysis.ipynb - Colaboratory

1

2

airline_sentiment = airline_tweets.groupby(['airline', 'airline_sentiment']).airline_se

airline_sentiment.plot(kind='bar')

<matplotlib.axes._subplots.AxesSubplot at 0x7f7545b94a50>

1

2

3

features = airline_tweets.iloc[:, 10].values # Text shared

labels = airline_tweets.iloc[:, 1].values # Airline Sentiment (Neutral, Negative & Posi

print(features)

['@VirginAmerica What @dhepburn said.'

"@VirginAmerica plus you've added commercials to the experience... tacky."

"@VirginAmerica I didn't today... Must mean I need to take another trip!"

... '@AmericanAir Please bring American Airlines to #BlackBerry10'

"@AmericanAir you have my money, you change my flight, and don't answer your phones

'@AmericanAir we have 8 ppl so we need 2 know how many seats are on the next flight

1

2

import nltk

nltk.download('stopwords')

[nltk_data] Downloading package stopwords to /root/nltk_data...

[nltk_data] Unzipping corpora/stopwords.zip.

True

1 print(labels)

https://colab.research.google.com/drive/1SBn5i_qSYKVq0ThDZh0ZG2WC2-C7czEx#printMode=true

4/7

13/11/2021 10:13

TF-IDF for Sentiment Analysis.ipynb - Colaboratory

['neutral' 'positive' 'neutral' ... 'neutral' 'negative' 'neutral']

1

2

3

from nltk.corpus import stopwords

from sklearn.feature_extraction.text import TfidfVectorizer

tidy_features = []

1

2

3

4

5

6

7

8

9

10

11

12

13

14

15

16

17

18

19

20

print (" Features **")

print(features)

for sentence in range(0, len(features)):

Remove all the special characters

tidy_feature = re.sub(r'\W', ' ', str(features[sentence]))

remove all single characters: espace + Charcater + Espace Bonjour a my

tidy_feature= re.sub(r'\s+[a-zA-Z]\s+', ' ', tidy_feature)

remove all single characters: espace + Charcater + Espace Bonjour a my

tidy_feature= re.sub(r'\s+[^0-9]\s+', ' ', tidy_feature)

Substituting multiple spaces with single space

tidy_feature = re.sub(r'\s+', ' ', tidy_feature, flags=re.I)

Subtituting a prefexied @ with ''

tidy_feature = re.sub(r'^@\s+', '', tidy_feature)

Subtituting a prefexied @ with ''

tidy_feature = re.sub(r'\d+', '', tidy_feature)

tidy_feature = tidy_feature.lower()

tidy_features.append(tidy_feature)

print (" Tidy Features **")

print(tidy_features)

Features **

['@VirginAmerica What @dhepburn said.'

"@VirginAmerica plus you've added commercials to the experience... tacky."

"@VirginAmerica I didn't today... Must mean I need to take another trip!"

... '@AmericanAir Please bring American Airlines to #BlackBerry10'

"@AmericanAir you have my money, you change my flight, and don't answer your phones

'@AmericanAir we have 8 ppl so we need 2 know how many seats are on the next flight

Tidy Features **

[' virginamerica what dhepburn said ', ' virginamerica plus you ve added commercials

1

2

3

vectorizer = TfidfVectorizer (max_features=2500, min_df=7, max_df=0.8, stop_words=stopw

numerical_tidy_features = vectorizer.fit_transform(tidy_features).toarray()

print(numerical_tidy_features)

[[0. 0. 0. ... 0. 0. 0.]

[0. 0. 0. ... 0. 0. 0.]

[0. 0. 0. ... 0. 0. 0.]

...

[0. 0. 0. ... 0. 0. 0.]

[0. 0. 0. ... 0. 0. 0.]

[0. 0. 0. ... 0. 0. 0.]]

1

2

from sklearn.model_selection import train_test_split

X_train, X_test, y_train, y_test = train_test_split(numerical_tidy_features, labels, te

https://colab.research.google.com/drive/1SBn5i_qSYKVq0ThDZh0ZG2WC2-C7czEx#printMode=true

5/7

13/11/2021 10:13

TF-IDF for Sentiment Analysis.ipynb - Colaboratory

1 print(type(y_train))

<class 'numpy.ndarray'>

1

2

3

from sklearn.svm import SVC

text_classifier = SVC(kernel='rbf')

text_classifier.fit(X_train, y_train)

SVC(C=1.0, break_ties=False, cache_size=200, class_weight=None, coef0=0.0,

decision_function_shape='ovr', degree=3, gamma='scale', kernel='rbf',

max_iter=-1, probability=False, random_state=None, shrinking=True,

tol=0.001, verbose=False)

1 predictions = text_classifier.predict(X_test)

1

2

3

4

from sklearn.metrics import classification_report, confusion_matrix, accuracy_score

print(confusion_matrix(y_test,predictions))

print(classification_report(y_test,predictions))

print(accuracy_score(y_test, predictions))

[[609 452 809]

[ 62 207 345]

[ 26 52 366]]

precision recall f1-score support

negative 0.87 0.33 0.47 1870

neutral 0.29 0.34 0.31 614

positive 0.24 0.82 0.37 444

accuracy 0.40 2928

macro avg 0.47 0.50 0.39 2928

weighted avg 0.66 0.40 0.43 2928

0.4036885245901639

1

https://colab.research.google.com/drive/1SBn5i_qSYKVq0ThDZh0ZG2WC2-C7czEx#printMode=true

6/7

13/11/2021 10:13

TF-IDF for Sentiment Analysis.ipynb - Colaboratory

https://colab.research.google.com/drive/1SBn5i_qSYKVq0ThDZh0ZG2WC2-C7czEx#printMode=true

7/7