problem_solving/Problem13.py at master · joel1794/problem_solving · GitHub

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
import numpy as np
from numpy import linalg

malware = np.array([[1, -1, 1, -1, -1, 1], [-2, 2, 2, -1, -2, 2], [1, 3, 0, 1, 3, 1], [2, 3, 1, 1, -2, 0]])
benign = np.array([[-1, 2, 1, 2, -1, 0], [-2, 1, 2, 3, 2, 1], [-1, 3, 0, 1, 3, -1], [0, 2, 3, 1, 1, -2]])
test = np.array([[1, 5, 1, 5, 5, 1], [-2, 3, 2, 3, 0, 2], [2, -3, 2, 3, 0, 0], [2, -2, 2, 2, -1, 1]])

unit_eigenvectors = np.array([[0.16408622, 0.62780739, -0.26039808, -0.53891957, 0.46372117, 0.0751984],
                     [0.2443382, 0.1070203, -0.80167357, 0.42774159, -0.1373179, -0.29042369]])
training_delta = np.array([[-1.1069, 1.2793, -2.6800, 2.5076], [1.5480, 0.5484, -1.2085, -0.8879]])

malware = malware.transpose()
benign = benign.transpose()
test = test.transpose()
unit_eigenvectors = unit_eigenvectors.transpose()

# print(malware, "\n", benign, "\n", test)

# mean = [0] * len(malware)
# print(XT)
# for i in range(len(malware)):
#     mean[i] = sum(malware[i, :])/len(malware[i, :])

mean = np.array([7/4, 7/4, 5/4, 2, 2, 1])
malware = malware.astype(float)
benign = benign.astype(float)
test = test.astype(float)
# mean = np.array(mean)
print("mean to be subtracted is\n", mean)
for i in range(len(malware)):
    malware[i, :] = np.subtract(malware[i, :], mean[i])

for i in range(len(benign)):
    benign[i, :] = np.subtract(benign[i, :], mean[i])

for i in range(len(test)):
    test[i, :] = np.subtract(test[i, :], mean[i])

print("malware data after subtracting mean is\n", malware)
print("benign data after subtracting mean is\n", benign)
print("test data after subtracting mean is\n", test)
# covariance_matrix = np.dot(malware, malware.transpose())
# # print(covariance_matrix)
# covariance_matrix /= len(malware[0])
# print("covariance matrix of malware is\n", covariance_matrix)
# lambdas, U = linalg.eig(covariance_matrix)
#
# print("lambdas are\n", lambdas.round(3))
# print("U is\n", U.round(3))
#
# lambdas = lambdas[0:3]
# U = U[:, 0:3]
# U = np.dot(U, -1)

# print("lambdas are\n", lambdas.round(3))
# print("U is\n", U.round(3))

W_malware = []
W_benign = []
W_test = []

for j in range(len(training_delta)):
    W_malware.append([0] * len(training_delta[0]))
    W_benign.append([0] * len(training_delta[0]))
    W_test.append([0] * len(training_delta[0]))

W_malware = np.array(W_malware)
W_malware = W_malware.astype(float)
W_benign = np.array(W_benign)
W_benign = W_benign.astype(float)
W_test = np.array(W_test)
W_test = W_test.astype(float)
# print("delta is\n", delta)

# print("malware after subtracting mean is\n", malware)
# print("benign samples after subtracting mean is\n", benign)
# print("test samples after subtracting mean is\n", test)
# print("unit eigen vectors are\n", unit_eigenvectors)

for i in range(len(unit_eigenvectors[0])):
    for j in range(len(malware[0])):
        # print(malware[:, j], unit_eigenvectors[:, i])
        W_malware[i][j] = np.dot(malware[:, j], unit_eigenvectors[:, i])
        # print(W_malware)

for i in range(len(unit_eigenvectors[0])):
    for j in range(len(benign[0])):
        # print(benign[:, j], unit_eigenvectors[:, i])
        W_benign[i][j] = np.dot(benign[:, j], unit_eigenvectors[:, i])
        # print(W_benign)

for i in range(len(unit_eigenvectors[0])):
    for j in range(len(test[0])):
        # print(benign[:, j], unit_eigenvectors[:, i])
        W_test[i][j] = np.dot(test[:, j], unit_eigenvectors[:, i])
        # print(W_benign)

# print("training delta is\n", training_delta.round(2))
print("W malware is\n", W_malware.round(2))
print("W benign is\n", W_benign.round(2))
print("W test is\n", W_test.round(2))

min_delta_test = []
for i in range(len(W_test[0])):
    min_dist_malware = 99999
    min_dist_benign = 99999

    for j in range(len(W_malware[0])):
        # print(training_delta[:, j], W_malware[:, i])
        euc_dist = np.linalg.norm(W_malware[:, j] - W_test[:, i])
        # print(euc_dist.round(2))
        if euc_dist < min_dist_malware:
            min_dist_malware = euc_dist
            # print(training_delta[:, j], W_malware[:, i].round(2))
            # print(euc_dist.round(2))

    for j in range(len(W_benign[0])):
        # print(training_delta[:, j], W_malware[:, i])
        euc_dist = np.linalg.norm(W_benign[:, j] - W_test[:, i])
        # print(euc_dist.round(2))
        if euc_dist < min_dist_benign:
            min_dist_benign = euc_dist
            # print(training_delta[:, j], W_malware[:, i].round(2))
            # print(euc_dist.round(2))
    if min_dist_malware < min_dist_benign:
        print("nearest euclidean distance with malware sample is ", min_dist_malware.round(2))
        print("Y", i + 1, " is malware")
    else:
        print("nearest euclidean distance with benign sample is ", min_dist_benign.round(2))
        print("Y", i + 1, " is benign")
    # print('{0:.2f}'.format(min_dist))