-
Notifications
You must be signed in to change notification settings - Fork 0
Expand file tree
/
Copy pathspamclassifier.py
More file actions
148 lines (119 loc) · 4.49 KB
/
spamclassifier.py
File metadata and controls
148 lines (119 loc) · 4.49 KB
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
#data comes from http://spamassassin.apache.org/old/publiccorpus/
import os
import random
import re
import email
import numpy as np
from sklearn.model_selection import train_test_split
import tensorflow as tf
from tensorflow import keras
spam_path = 'input/spam_2/'
easy_ham_path = 'input/easy_ham/'
easy_ham_2_path = 'input/easy_ham_2/'
hard_ham_path = 'input/hard_ham/'
# label messages according to folder
email_files = {'spam': os.listdir(spam_path),
'easy_ham': os.listdir(easy_ham_path),
'easy_ham_2': os.listdir(easy_ham_2_path),
'hard_ham': os.listdir(hard_ham_path)
}
#count number of messages for each folder
print('\n spam emails: %d \n \
easy ham emails: %d \n \
easy ham emails 2: %d \n \
hard ham emails: %d \n' %(len(email_files['spam']), len(email_files['easy_ham']), len(email_files['easy_ham_2']), len(email_files['hard_ham']))
)
#show one email
path = spam_path
filename = random.sample(os.listdir(path), 1)[0]
file = open(path + filename,'r',errors='ignore')
content = file.read()
print(content)
#just the body
print('----------------------------------------')
msg = email.message_from_string(content)
if msg.is_multipart():
body = []
for payload in msg.get_payload():
# if payload.is_multipart(): ...
body.append(payload.get_payload())
body = ' '.join(body)
else:
body = msg.get_payload()
print(body)
#preprocessing
print('----------------------------------------')
raw_data = []
labels = []
invalid_list = []
def processemail(body):
body_pp = body.lower()
body_pp = re.sub(r"<[^<>]+>", " html ", body_pp)
body_pp = re.sub(r"[0-9]+", " number ", body_pp)
body_pp = re.sub(r"(http|https)://[^\s]*", ' httpaddr ', body_pp)
body_pp = re.sub(r"[^\s]+@[^\s]+", ' emailaddr ', body_pp)
body_pp = re.sub(r"[$]+", ' dollar ', body_pp)
body_pp = re.sub(r"[^a-zA-Z0-9]",' ', body_pp)
return body_pp
def processfolder(path, label):
for filename in os.listdir(path):
#print(filename)
try:
file = open(path + filename,'r',errors='ignore')
content = file.read()
msg = email.message_from_string(content)
if msg.is_multipart():
body = []
for payload in msg.get_payload():
# if payload.is_multipart(): ...
body.append(payload.get_payload())
body = ' '.join(body)
else:
body = msg.get_payload()
body = processemail(body)
raw_data.append(body)
labels.append(label)
except:
invalid_list.append(filename)
processfolder(spam_path, 1)
processfolder(easy_ham_path,0)
processfolder(easy_ham_2_path,0)
processfolder(hard_ham_path,0)
print("Total email count:{}".format(len(raw_data)))
print("Total labels: {}".format(len(labels)))
X_train_raw, X_test_raw, y_train, y_test = train_test_split(raw_data, labels, shuffle=True, test_size=0.33, random_state=42)
tokenizer = keras.preprocessing.text.Tokenizer(num_words=4096)
tokenizer.fit_on_texts(X_train_raw)
#convert the words to token sequences
X_train = tokenizer.texts_to_sequences(X_train_raw)
X_test = tokenizer.texts_to_sequences(X_test_raw)
#pad the sequences
X_train = keras.preprocessing.sequence.pad_sequences(X_train, value=0, padding='post', maxlen=2048)
X_test = keras.preprocessing.sequence.pad_sequences(X_test, value=0, padding='post', maxlen=2048)
print("Train size:{}".format(len(X_train)))
print("Test size:{}".format(len(X_test)))
model = keras.Sequential()
model.add(keras.layers.Embedding(4096, 16))
model.add(keras.layers.GlobalAveragePooling1D())
model.add(keras.layers.Dense(16, activation=tf.nn.relu))
model.add(keras.layers.Dense(16, activation=tf.nn.relu))
model.add(keras.layers.Dense(1, activation=tf.nn.sigmoid))
print(model.summary())
model.compile(optimizer=tf.train.AdamOptimizer(),
loss='binary_crossentropy',
metrics=['accuracy'])
#get a validation set
x_val = X_train[:691]
partial_x_train = X_train[691:]
y_val = y_train[:691]
partial_y_train = y_train[691:]
early_stop = keras.callbacks.EarlyStopping(monitor='val_loss', patience=15)
history = model.fit(partial_x_train,
partial_y_train,
epochs=300,
batch_size=100,
validation_data=(x_val, y_val),
verbose=1, callbacks=[early_stop])
results = model.evaluate(X_test, y_test)
print("Final Test Set Results: {}".format(results))
#Results are ~98%