-
Notifications
You must be signed in to change notification settings - Fork 2
/
Copy pathhyperpart_news_det_NB_v0.2.py
200 lines (157 loc) · 10 KB
/
hyperpart_news_det_NB_v0.2.py
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
# Hyperpartisan news detection- exploratory analyses and model selection- Naive Bayes approaches v0.2
# Author: Yash Deshpande
# Last modified: December 2nd, 2018
# Updates:
# 1. Added all naive Bayes' approaches, including CNB
# 2. Working with fully cleaned data
import datetime
print("Naive Bayes script started at " + str(datetime.datetime.now()))
import pandas as pd
import string
import numpy as np
from sklearn import model_selection as ms
from sklearn.feature_extraction.text import TfidfVectorizer
import matplotlib.pyplot as plt
from sklearn.naive_bayes import BernoulliNB
from sklearn.naive_bayes import MultinomialNB
from sklearn.naive_bayes import ComplementNB
from sklearn import metrics
# Loading data and mapping character classes to numerics
training_data_df = pd.read_csv('sample_data_labelled_cleaned.csv.gz', sep = '|', header = 0)
training_data_df['Hyperpartisan'][training_data_df['Hyperpartisan'] == True] = 1
training_data_df['Hyperpartisan'][training_data_df['Hyperpartisan'] == False] = 0
X = training_data_df['Article']
Y = training_data_df['Hyperpartisan']
Y = Y.astype(int)
X_train, X_test, Y_train, Y_test = ms.train_test_split(X, Y, test_size = 0.25)
# Vectorizing inputs using TF-IDF
ti_vectorize_data_12 = TfidfVectorizer(ngram_range=(1, 2), stop_words = 'english').fit(X)
x_train_vec_tfidf_12, x_test_vec_tfidf_12 = ti_vectorize_data_12.transform(X_train), ti_vectorize_data_12.transform(X_test)
ti_vectorize_data_14 = TfidfVectorizer(ngram_range=(1, 4), stop_words = 'english').fit(X)
x_train_vec_tfidf_14, x_test_vec_tfidf_14 = ti_vectorize_data_14.transform(X_train), ti_vectorize_data_14.transform(X_test)
ti_vectorize_data_16 = TfidfVectorizer(ngram_range=(1, 6), stop_words = 'english').fit(X)
x_train_vec_tfidf_16, x_test_vec_tfidf_16 = ti_vectorize_data_16.transform(X_train), ti_vectorize_data_16.transform(X_test)
ti_vectorize_data_610 = TfidfVectorizer(ngram_range=(6, 10), stop_words = 'english').fit(X)
x_train_vec_tfidf_610, x_test_vec_tfidf_610 = ti_vectorize_data_610.transform(X_train), ti_vectorize_data_610.transform(X_test)
# Bernoulli Naive Bayes models
BNB_model_ti_12 = BernoulliNB().fit(x_train_vec_tfidf_12, Y_train)
preds_BNB_tfidf_12 = BNB_model_ti_12.predict_proba(x_test_vec_tfidf_12)
accuracy_BNB_12 = metrics.accuracy_score(Y_test, BNB_model_ti_12.predict(x_test_vec_tfidf_12))
BNB_model_ti_14 = BernoulliNB().fit(x_train_vec_tfidf_14, Y_train)
preds_BNB_tfidf_14 = BNB_model_ti_14.predict_proba(x_test_vec_tfidf_14)
accuracy_BNB_14 = metrics.accuracy_score(Y_test, BNB_model_ti_14.predict(x_test_vec_tfidf_14))
BNB_model_ti_16 = BernoulliNB().fit(x_train_vec_tfidf_16, Y_train)
preds_BNB_tfidf_16 = BNB_model_ti_16.predict_proba(x_test_vec_tfidf_16)
accuracy_BNB_16 = metrics.accuracy_score(Y_test, BNB_model_ti_16.predict(x_test_vec_tfidf_16))
BNB_model_ti_610 = BernoulliNB().fit(x_train_vec_tfidf_610, Y_train)
preds_BNB_tfidf_610 = BNB_model_ti_610.predict_proba(x_test_vec_tfidf_610)
accuracy_BNB_610 = metrics.accuracy_score(Y_test, BNB_model_ti_610.predict(x_test_vec_tfidf_610))
# Multinomial Naive Bayes models
MNB_model_ti_12 = MultinomialNB().fit(x_train_vec_tfidf_12, Y_train)
preds_MNB_tfidf_12 = MNB_model_ti_12.predict_proba(x_test_vec_tfidf_12)
accuracy_MNB_12 = metrics.accuracy_score(Y_test, BNB_model_ti_12.predict(x_test_vec_tfidf_12))
MNB_model_ti_14 = MultinomialNB().fit(x_train_vec_tfidf_14, Y_train)
preds_MNB_tfidf_14 = MNB_model_ti_14.predict_proba(x_test_vec_tfidf_14)
accuracy_MNB_14 = metrics.accuracy_score(Y_test, BNB_model_ti_14.predict(x_test_vec_tfidf_14))
MNB_model_ti_16 = MultinomialNB().fit(x_train_vec_tfidf_16, Y_train)
preds_MNB_tfidf_16 = MNB_model_ti_16.predict_proba(x_test_vec_tfidf_16)
accuracy_MNB_16 = metrics.accuracy_score(Y_test, BNB_model_ti_16.predict(x_test_vec_tfidf_16))
MNB_model_ti_610 = MultinomialNB().fit(x_train_vec_tfidf_610, Y_train)
preds_MNB_tfidf_610 = MNB_model_ti_610.predict_proba(x_test_vec_tfidf_610)
accuracy_MNB_610 = metrics.accuracy_score(Y_test, BNB_model_ti_610.predict(x_test_vec_tfidf_610))
# Complement Naive Bayes models
CNB_model_ti_12 = ComplementNB().fit(x_train_vec_tfidf_12, Y_train)
preds_CNB_tfidf_12 = CNB_model_ti_12.predict_proba(x_test_vec_tfidf_12)
accuracy_CNB_12 = metrics.accuracy_score(Y_test, CNB_model_ti_12.predict(x_test_vec_tfidf_12))
CNB_model_ti_14 = ComplementNB().fit(x_train_vec_tfidf_14, Y_train)
preds_CNB_tfidf_14 = CNB_model_ti_14.predict_proba(x_test_vec_tfidf_14)
accuracy_CNB_14 = metrics.accuracy_score(Y_test, CNB_model_ti_14.predict(x_test_vec_tfidf_14))
CNB_model_ti_16 = ComplementNB().fit(x_train_vec_tfidf_16, Y_train)
preds_CNB_tfidf_16 = CNB_model_ti_16.predict_proba(x_test_vec_tfidf_16)
accuracy_CNB_16 = metrics.accuracy_score(Y_test, CNB_model_ti_16.predict(x_test_vec_tfidf_16))
CNB_model_ti_610 = ComplementNB().fit(x_train_vec_tfidf_610, Y_train)
preds_CNB_tfidf_610 = CNB_model_ti_610.predict_proba(x_test_vec_tfidf_610)
accuracy_CNB_610 = metrics.accuracy_score(Y_test, CNB_model_ti_610.predict(x_test_vec_tfidf_610))
# Accuracies DF
accuracies_list = [['BNB_12', accuracy_BNB_12],
['BNB_14', accuracy_BNB_14],
['BNB_16', accuracy_BNB_16],
['BNB_610', accuracy_BNB_610],
['MNB_12', accuracy_MNB_12],
['MNB_14', accuracy_MNB_14],
['MNB_16', accuracy_MNB_16],
['MNB_610', accuracy_MNB_610],
['CNB_12', accuracy_CNB_12],
['CNB_14', accuracy_CNB_14],
['CNB_16', accuracy_CNB_16],
['CNB_610', accuracy_CNB_610]]
NB_accuracies_df = pd.DataFrame(accuracies_list)
NB_accuracies_df.columns = ['Model', 'Accuracy']
NB_accuracies_df.to_csv("NB_accuracies_df.csv", sep = '|')
# FPR, TPR calculations for all models
fpr_BNB_ti_12, tpr_BNB_ti_12, thresholds_BNB_ti_12 = metrics.roc_curve(Y_test, preds_BNB_tfidf_12[:,1])
fpr_BNB_ti_14, tpr_BNB_ti_14, thresholds_BNB_ti_14 = metrics.roc_curve(Y_test, preds_BNB_tfidf_14[:,1])
fpr_BNB_ti_16, tpr_BNB_ti_16, thresholds_BNB_ti_16 = metrics.roc_curve(Y_test, preds_BNB_tfidf_16[:,1])
fpr_BNB_ti_610, tpr_BNB_ti_610, thresholds_BNB_ti_610 = metrics.roc_curve(Y_test, preds_BNB_tfidf_610[:,1])
fpr_MNB_ti_12, tpr_MNB_ti_12, thresholds_MNB_ti_12 = metrics.roc_curve(Y_test, preds_MNB_tfidf_12[:,1])
fpr_MNB_ti_14, tpr_MNB_ti_14, thresholds_MNB_ti_14 = metrics.roc_curve(Y_test, preds_MNB_tfidf_14[:,1])
fpr_MNB_ti_16, tpr_MNB_ti_16, thresholds_MNB_ti_16 = metrics.roc_curve(Y_test, preds_MNB_tfidf_16[:,1])
fpr_MNB_ti_610, tpr_MNB_ti_610, thresholds_MNB_ti_610 = metrics.roc_curve(Y_test, preds_MNB_tfidf_610[:,1])
fpr_CNB_ti_12, tpr_CNB_ti_12, thresholds_CNB_ti_12 = metrics.roc_curve(Y_test, preds_CNB_tfidf_12[:,1])
fpr_CNB_ti_14, tpr_CNB_ti_14, thresholds_CNB_ti_14 = metrics.roc_curve(Y_test, preds_CNB_tfidf_14[:,1])
fpr_CNB_ti_16, tpr_CNB_ti_16, thresholds_CNB_ti_16 = metrics.roc_curve(Y_test, preds_CNB_tfidf_16[:,1])
fpr_CNB_ti_610, tpr_CNB_ti_610, thresholds_CNB_ti_610 = metrics.roc_curve(Y_test, preds_CNB_tfidf_610[:,1])
# AUC calculations for all models
roc_auc_BNB_ti_12 = metrics.auc(fpr_BNB_ti_12, tpr_BNB_ti_12)
roc_auc_BNB_ti_14 = metrics.auc(fpr_BNB_ti_14, tpr_BNB_ti_14)
roc_auc_BNB_ti_16 = metrics.auc(fpr_BNB_ti_16, tpr_BNB_ti_16)
roc_auc_BNB_ti_610 = metrics.auc(fpr_BNB_ti_610, tpr_BNB_ti_610)
roc_auc_MNB_ti_12 = metrics.auc(fpr_MNB_ti_12, tpr_MNB_ti_12)
roc_auc_MNB_ti_14 = metrics.auc(fpr_MNB_ti_14, tpr_MNB_ti_14)
roc_auc_MNB_ti_16 = metrics.auc(fpr_MNB_ti_16, tpr_MNB_ti_16)
roc_auc_MNB_ti_610 = metrics.auc(fpr_MNB_ti_610, tpr_MNB_ti_610)
roc_auc_CNB_ti_12 = metrics.auc(fpr_CNB_ti_12, tpr_CNB_ti_12)
roc_auc_CNB_ti_14 = metrics.auc(fpr_CNB_ti_14, tpr_CNB_ti_14)
roc_auc_CNB_ti_16 = metrics.auc(fpr_CNB_ti_16, tpr_CNB_ti_16)
roc_auc_CNB_ti_610 = metrics.auc(fpr_CNB_ti_610, tpr_CNB_ti_610)
# Plotting ROC curves
# BNB
plt.figure(figsize = (20, 10))
plt.title("ROC curves- Bernoulli Naive Bayes for various n-gram ranges", fontsize = 20)
plt.ylabel("True positive rate (TPR)", size = 15)
plt.xlabel("False positive rate (FPR)", size = 15)
plt.plot(fpr_BNB_ti_12, tpr_BNB_ti_12, 'y', c='r', label = 'n-gram range (1, 2)' +' (AUC = %0.3f)' % roc_auc_BNB_ti_12)
plt.plot(fpr_BNB_ti_14, tpr_BNB_ti_14, 'y', c='g', label = 'n-gram range (1, 4)' +' (AUC = %0.3f)' % roc_auc_BNB_ti_14)
plt.plot(fpr_BNB_ti_16, tpr_BNB_ti_16, 'y', c='b', label = 'n-gram range (1, 6)' +' (AUC = %0.3f)' % roc_auc_BNB_ti_16)
plt.plot(fpr_BNB_ti_610, tpr_BNB_ti_610, 'y', c='k', label = 'n-gram range (6, 10)' +' (AUC = %0.3f)' % roc_auc_BNB_ti_610)
x = np.linspace(0, 1, 5)
plt.plot(x, x, 'k--')
plt.legend(loc = "lower right")
plt.savefig("MS_plot_1_BNB.jpg")
# MNB
plt.figure(figsize = (20, 10))
plt.title("ROC curves- Multinomial Naive Bayes for various n-gram ranges", fontsize = 20)
plt.ylabel("True positive rate (TPR)", size = 15)
plt.xlabel("False positive rate (FPR)", size = 15)
plt.plot(fpr_MNB_ti_12, tpr_MNB_ti_12, 'y', c='r', label = 'n-gram range (1, 2)' +' (AUC = %0.3f)' % roc_auc_MNB_ti_12)
plt.plot(fpr_MNB_ti_14, tpr_MNB_ti_14, 'y', c='g', label = 'n-gram range (1, 4)' +' (AUC = %0.3f)' % roc_auc_MNB_ti_14)
plt.plot(fpr_MNB_ti_16, tpr_MNB_ti_16, 'y', c='b', label = 'n-gram range (1, 6)' +' (AUC = %0.3f)' % roc_auc_MNB_ti_16)
plt.plot(fpr_MNB_ti_610, tpr_MNB_ti_610, 'y', c='k', label = 'n-gram range (6, 10)' +' (AUC = %0.3f)' % roc_auc_MNB_ti_610)
x = np.linspace(0, 1, 5)
plt.plot(x, x, 'k--')
plt.legend(loc = "lower right")
plt.savefig("MS_plot_2_MNB.jpg")
# CNB
plt.figure(figsize = (20, 10))
plt.title("ROC curves- Complement Naive Bayes for various n-gram ranges", fontsize = 20)
plt.ylabel("True positive rate (TPR)", size = 15)
plt.xlabel("False positive rate (FPR)", size = 15)
plt.plot(fpr_CNB_ti_12, tpr_CNB_ti_12, 'y', c='r', label = 'n-gram range (1, 2)' +' (AUC = %0.3f)' % roc_auc_CNB_ti_12)
plt.plot(fpr_CNB_ti_14, tpr_CNB_ti_14, 'y', c='g', label = 'n-gram range (1, 4)' +' (AUC = %0.3f)' % roc_auc_CNB_ti_14)
plt.plot(fpr_CNB_ti_16, tpr_CNB_ti_16, 'y', c='b', label = 'n-gram range (1, 6)' +' (AUC = %0.3f)' % roc_auc_CNB_ti_16)
plt.plot(fpr_CNB_ti_610, tpr_CNB_ti_610, 'y', c='k', label = 'n-gram range (6, 10)' +' (AUC = %0.3f)' % roc_auc_CNB_ti_610)
x = np.linspace(0, 1, 5)
plt.plot(x, x, 'k--')
plt.legend(loc = "lower right")
plt.savefig("MS_plot_3_CNB.jpg")
print("Naive Bayes script completed at " + str(datetime.datetime.now()))