-
Notifications
You must be signed in to change notification settings - Fork 0
/
Copy pathhomework.py
201 lines (182 loc) · 7.34 KB
/
homework.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
from os import listdir
from os.path import isfile, join
from sklearn.metrics import confusion_matrix
from features_analyzer import analyzer
from time import sleep
import sys
mypath = '../drebin/feature_vectors'
allfiles = [f for f in listdir(mypath) if isfile(join(mypath, f))]
# Print iterations progress
def printProgressBar (iteration, total, prefix = '', suffix = '', decimals = 1, length = 50, fill = '█'):
percent = ("{0:." + str(decimals) + "f}").format(100 * (iteration / float(total)))
filledLength = int(length * iteration // total)
bar = fill * filledLength + '-' * (length - filledLength)
print('\r%s |%s| %s%% %s' % (prefix, bar, percent, suffix), end = '\r')
# Print New Line on Complete
if iteration == total:
print()
lista_malware = list()
def load_malwares():
with open('../drebin/sha256_family.csv', 'r') as f:
raw_data = f.read()
f.close()
raw_data = raw_data.strip()
lista_data = raw_data.split('\n')
for i in range(len(lista_data)):
if i != 0:
lista_malware.append((lista_data[i].split(','))[0])
return lista_malware
def train(lista_file):
#Training dei dati
mal,nomal = dict(),dict()
n_mal,n_nomal = 0,0
printProgressBar(0, len(lista_file), prefix = 'Training:', suffix = 'Complete')
for a in range(len(lista_file)):
printProgressBar(a+1, len(lista_file), prefix = 'Training:', suffix = 'Complete')
path = '../drebin/feature_vectors/'+lista_file[a]
with open(path, 'r') as f:
raw_data = f.read()
f.close()
raw_data = raw_data.strip()
#data.append(raw_data.split('\n'))
data = raw_data.split('\n')
for e in range(len(data)):
if(lista_file[a] in lista_malware):
n_mal += 1
if data[e] not in mal.keys():
mal[data[e]] = 0
else:
mal[data[e]] +=1
else:
n_nomal += 1
if data[e] not in nomal.keys():
nomal[data[e]] = 0
else:
nomal[data[e]] +=1
return n_mal,n_nomal,mal,nomal
def bernoulli_naive(doc, mal, nomal, n_mal, n_nomal, dic):
p_mal, p_nomal = 1.0, 1.0
#print(dic)
for s in dic:
if s in mal.keys():
p_m = (mal[s]+1)/(n_mal+2) #probabilita di trovare quella parola in mal
else:
p_m = 1/(n_mal+2)
if s in nomal.keys():
p_n = (nomal[s]+1)/(n_nomal+2) #probabilita di trovare quella parola in nomal
else:
p_n = 1/(n_nomal+2)
if s in doc:
p_mal *= p_m #0.09 #andiamo a trovare la probabilita che quella parola s della nostra frase si trovi in mal
p_nomal *= p_n #0.01
else:
p_mal *= (1-p_m) #0.91 #probabilita di ... e 1 meno la probabilita di trovare quella parola in mal
p_nomal *= (1-p_n) #0.99
t = max(p_mal, p_nomal)
if t==p_mal:
return 'mal'
elif t==p_nomal:
return 'nomal'
if __name__ == "__main__":
lista_malware = load_malwares()
data = list()
dic = list()
tipo = input('What dictionary to be chosen? 1 (All the features) o 2 (top 8 features): ')
if tipo == '1':
print("Initializing dictionary")
printProgressBar(0, len(allfiles), prefix = 'Progress:', suffix = 'Complete')
for counter,a in enumerate(allfiles):
path = '../drebin/feature_vectors/'+a
with open(path, 'r') as f:
raw_data = f.read()
f.close()
raw_data = raw_data.strip()
data = raw_data.split('\n')
for i in range(len(data)-1):
dic.append(data[i]) #senza splittare la frase
printProgressBar(counter + 1, len(allfiles), prefix = 'Progress:', suffix = 'Complete')
dic = set(dic)
print("Dictionary done")
else:
dic = analyzer(lista_malware)
acc = list()
train_data = allfiles[0:((len(allfiles)*80)//100)]
test_data = allfiles[len(train_data):len(allfiles)]
y_pred,y_true = list(),list()
n_mal,n_nomal,mal,nomal = train(train_data)
for t in range(len(test_data)):
printProgressBar(t+1, len(test_data), prefix = 'Testing:', suffix = 'Complete')
lista_value = list()
path = '../drebin/feature_vectors/'+test_data[t]
if test_data[t] in lista_malware:
truth = 'mal'
else:
truth = 'nomal'
with open(path, 'r') as f:
raw_data = f.read()
f.close()
raw_data = raw_data.strip()
data = raw_data.split('\n')
for i in data:
lista_value.append(i)
c_map = bernoulli_naive(lista_value,mal,nomal,n_mal,n_nomal,dic)
y_pred.append(c_map)
y_true.append(truth)
if c_map == truth:
acc.append(1)
else:
acc.append(0)
print('Accuracy is: {}'.format(sum(acc)/len(acc)))
conf_matrix = confusion_matrix(y_true, y_pred, labels=["mal", "nomal"])
print('Confusion matrix is: \n {}'.format(conf_matrix))
TN, FP, FN, TP = confusion_matrix(y_true, y_pred, labels=["mal", "nomal"]).ravel()
precision = TP/(TP+FP)
recall = TP/(TP+FN)
print("Precision: {}".format(precision))
print("Recall: {}".format(recall))
print("False Positive Rate: {}".format(FP/(FP+TN)))
print("Accuracy: {}".format(((TP+TN)/(TP+FN+TN+FP))))
print("F-Measure: {}".format(2*(precision*recall)/(precision+recall)))
# Confusion matrix is:
# [[ 33 1092]
# [ 42 24636]]
# #k-fold cross validation
# fold_size = len(allfiles)//30
# acc = list()
# print('Lunghezza file: {}'.format(len(allfiles)))
# for i in range(0, len(allfiles), fold_size):
# test_data = allfiles[i:i+fold_size]
# train_data = allfiles[0:i] + allfiles[(i+fold_size):]
# print("Iniziando il training")
# n_mal,n_nomal,mal,nomal = train(train_data)
# print("Training completato")
# f_acc = list()
# printProgressBar(0, len(test_data), prefix = 'Testing:', suffix = 'Complete')
# for t in range(len(test_data)):
# printProgressBar(t+1, len(test_data), prefix = 'Testing:', suffix = 'Complete')
# lista_value = list()
# path = '../drebin/feature_vectors/'+test_data[t]
# if test_data[t] in lista_malware:
# truth = 'mal'
# else:
# truth = 'nomal'
# with open(path, 'r') as f:
# raw_data = f.read()
# f.close()
# raw_data = raw_data.strip()
# #data.append(raw_data.split('\n'))
# data = raw_data.split('\n')
# for i in data:
# # elemento_test = i.split('::')
# # lista_value.append(elemento_test[1])
# lista_value.append(i)
# c_map = bernoulli_naive(lista_value,mal,nomal,n_mal,n_nomal,dic)
# #print("Il programma ha rilevato che il file é: {} / {}".format(c_map,truth))
# if c_map == truth:
# f_acc.append(1)
# acc.append(1)
# else:
# f_acc.append(0)
# acc.append(0)
# print('Accuracy for k is: {}'.format(sum(f_acc)/len(f_acc)))
# print('Final Accuracy for fold size {} is: {}'.format(fold_size, sum(acc)/len(acc)))