-
-
Notifications
You must be signed in to change notification settings - Fork 0
/
main.py
207 lines (163 loc) · 7.29 KB
/
main.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
201
202
203
204
205
206
207
import random
from sklearn.neighbors import KNeighborsClassifier
from sklearn.ensemble import RandomForestClassifier
from sklearn import tree
from sklearn import svm
from sklearn.model_selection import cross_validate
import numpy as np
import time
import click
import feature_extraction
start_time = time.time()
X = []
Y = []
def store_features(path):
# saving final extracted features for probabilistic future use
file = open(path, 'w')
for i, x in enumerate(X):
for item in x:
file.write('{},'.format(str(item)))
file.write(str(Y[i][0]) + '\n')
file.close()
def load_features(path):
X = []
Y = []
file = open(path, 'r')
lines = file.readlines()
for i, line in enumerate(lines):
X.append([float(x) for x in line.split(',')[0:-1]])
Y.append(int(line.split(',')[-1]))
file.close()
return X, Y
def load_data(malwarepath, benignpath, benignheaderfieldspath, malwareheaderfieldspath, malwaresectionnamespath,
benignsectionnamespath):
file = open(malwareheaderfieldspath, 'r')
malware_header_fields = file.readlines()
file.close()
file = open(malwaresectionnamespath, 'r')
malware_section_names = file.readlines()
file.close()
file = open(benignheaderfieldspath, 'r')
benign_header_fields = file.readlines()
file.close()
file = open(benignsectionnamespath, 'r')
benign_section_names = file.readlines()
file.close()
return malwarepath, benignpath, benign_header_fields, malware_header_fields, benign_section_names, malware_section_names
def log(message):
print(message)
def final_features_extraction(path, header_fields, section_names, label):
for i, row in enumerate(header_fields):
final_features = []
Y.append([label])
row = row.split('\t,')
sample_name = row[-1].strip('\n')
# derived features
entropies = feature_extraction.entropy(sample_name, path)
final_features.append(entropies[0])
final_features.append(entropies[1])
final_features.append(entropies[2])
sectionnames = section_names[i]
sectionnames = sectionnames.split(',')
sectionnames.remove(sectionnames[-1])
section_name_features = feature_extraction.section_name_checker(sectionnames)
final_features.append(section_name_features[0])
final_features.append(section_name_features[1])
final_features.append(feature_extraction.compilation_time(row[21]))
final_features.append(feature_extraction.extract_file_size(sample_name, path))
final_features.append(feature_extraction.extract_file_info(sample_name, path))
final_features.append(feature_extraction.Image_Base_checker(row[34]))
final_features.append(feature_extraction.sectionalignment_checker(int(row[35]), int(row[36])))
final_features.append(feature_extraction.filealignment_checker(int(row[35]), int(row[36])))
final_features.append(feature_extraction.sizeofimage_checker(int(row[44]), int(row[35])))
final_features.append(feature_extraction.size_of_header_checker(sample_name, path))
# Expanded features
zerofill = bin(int(row[25]))[2:].zfill(16)
characteristics = zerofill[0:6] + zerofill[7:]
for c in characteristics:
final_features.append(c)
Dllzerofill = bin(int(row[48]))[2:].zfill(16)
dllcharacteristics = Dllzerofill[5:]
for d in dllcharacteristics:
final_features.append(d)
# raw features
final_features.append(row[0])
final_features.append(row[1])
final_features.append(row[2])
final_features.append(row[3])
final_features.append(row[4])
final_features.append(row[5])
final_features.append(row[19])
final_features.append(row[26])
final_features.append(row[27])
final_features.append(row[28])
final_features.append(row[29])
final_features.append(row[30])
final_features.append(row[31])
final_features.append(row[32])
final_features.append(row[33])
final_features.append(row[34])
final_features.append(row[35])
final_features.append(row[36])
final_features.append(row[37])
final_features.append(row[38])
final_features.append(row[39])
final_features.append(row[40])
final_features.append(row[41])
final_features.append(row[42])
final_features.append(row[43])
final_features.append(row[44])
final_features.append(row[45])
final_features.append(row[46])
X.append(final_features)
return X, Y
def learning(X, Y):
algorithms = {
"RandomForest": RandomForestClassifier(),
"SVM": svm.SVC(),
"Knn": KNeighborsClassifier(n_neighbors=5),
"DecisionTree": tree.DecisionTreeClassifier(),
}
for algo in algorithms:
print('{} results'.format(algo))
start_time = time.time()
clf = algorithms[algo]
scores = cross_validate(clf, X, Y, cv=10, scoring=('accuracy', 'f1', 'recall', 'precision'))
for score_name in ['test_accuracy', 'test_precision', 'test_recall', 'test_f1']:
print('{} : {}'.format(score_name, np.mean(scores[score_name])))
end_time = time.time()
execution_time = end_time - start_time
print('{} execution time {} \n'.format(algo, execution_time))
@click.command()
@click.option("--malwarepath", required=True, help="path of malware samples")
@click.option("--benignpath", required=True, help="path of benign samples")
@click.option("--benignheaderfieldspath", required=True, help="path of stored header fields file for benign samples")
@click.option("--malwareheaderfieldspath", required=True, help="path of stored header fields file for malware samples")
@click.option("--malwaresectionnamespath", required=True, help="path of stored header fields file for malware samples")
@click.option("--benignsectionnamespath", required=True, help="path of stored header fields file for malware samples")
def main(malwarepath, benignpath, benignheaderfieldspath, malwareheaderfieldspath, malwaresectionnamespath,
benignsectionnamespath):
malware_path, benign_path, benign_header_fields, malware_header_fields, benign_section_names, malware_section_names = \
load_data(malwarepath, benignpath, benignheaderfieldspath, malwareheaderfieldspath, malwaresectionnamespath,
benignsectionnamespath)
log("processing malwares for extracting features")
X, Y = final_features_extraction(malware_path, malware_header_fields, malware_section_names, 1)
log("processing benign samples for extracting features")
X, Y = final_features_extraction(benign_path, benign_header_fields, benign_section_names, 0)
global start_time
end_time = time.time()
feature_extraction_time = end_time - start_time
print('feature extraction time {}'.format(feature_extraction_time))
# saving final extracted features for probabilistic future use
store_features('final_features.txt')
# extracted features loading
X, Y = load_features('final_features.txt')
# shuffle
start_time = time.time()
features_label = list(zip(X, Y))
random.shuffle(features_label)
X, Y = zip(*features_label)
# learning
learning(X, Y)
if __name__ == '__main__':
main()