Network-Intrusion-Detection/create_data.py at master · animesh-s/Network-Intrusion-Detection · 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
import pandas as pd
import numpy as np
import math
import pdb
import random

LABELS = {'guess_passwd.': 6, 'nmap.': 17, 'loadmodule.': 2, 'rootkit.': 22, 'warezclient.': 20,
            'smurf.': 5, 'pod.': 7, 'neptune.': 4, 'normal.': 0, 'spy.': 21, 'ftp_write.': 12, 'phf.': 16,
            'portsweep.': 9, 'teardrop.': 8, 'buffer_overflow.': 1, 'land.': 11, 'imap.': 14, 'warezmaster.': 19,
            'perl.': 3, 'multihop.': 18, 'back.': 13, 'ipsweep.': 10, 'satan.': 15}

def dataset_count(N, training_percent, validation_percent):
  training_count = (int)(training_percent * N)
  validation_count = (int)(validation_percent * N)
  test_count = N - training_count - validation_count
  return training_count, validation_count, test_count

def partition_data(indices, training_count, validation_count, test_count):
  indices = np.random.permutation(indices)
  training_samples = random.sample(indices, training_count)
  indices = [index for index in indices if index not in training_samples]
  validation_samples = random.sample(indices, validation_count)
  indices = [index for index in indices if index not in (training_samples + validation_samples)]
  test_samples = random.sample(indices, test_count)
  return training_samples, validation_samples, test_samples

def create_dataset(samples, labels, training_indices, validation_indices, test_indices):
  training_samples, validation_samples, test_samples, training_labels, validation_labels, test_labels = [], [], [], [], [], []
  writer = open('../data/training_set.csv', 'w')
  label_writer = open('../data/training_labels.txt', 'w')
  for index in training_indices:
      writer.write(samples[index] + '\n')
      label_writer.write(str(labels[index]) + '\n')
  writer.close()
  label_writer.close()
  writer = open('../data/validation_set.csv', 'w')
  label_writer = open('../data/validation_labels.txt', 'w')
  for index in validation_indices:
      writer.write(samples[index] + '\n')
      label_writer.write(str(labels[index]) + '\n')
  writer.close()
  label_writer.close()
  writer = open('../data/test_set.csv', 'w')
  label_writer = open('../data/test_labels.txt', 'w')
  for index in test_indices:
      writer.write(samples[index] + '\n')
      label_writer.write(str(labels[index]) + '\n')
  writer.close()
  label_writer.close()

def get_label(label):
  return LABELS[label]

def get_dicts(dataset):
  dict_1, dict_2, dict_3= {}, {}, {}
  count1, count2, count3 = 0, 0, 0
  for sample in dataset:
    value = sample[1]
    if(value not in dict_1):
        count1 = count1 + 1
        dict_1[value] = count1
    value = sample[2]
    if(value not in dict_2):
        count2 = count2 + 1;
        dict_2[value] = count2
    value = sample[3]
    if(value not in dict_3):
        count3 = count3 + 1
        dict_3[value] = count3
  return dict_1, dict_2, dict_3

def main():
  smurf_indices = []
  neptune_indices = []
  normal_indices = []
  other_indices = []
  index = 0
  data = pd.read_csv('../data/kddcup.data_10_percent', sep=',',header=None).values
  for row in data:
    label = row[41]
    if label == "normal.":
      normal_indices.append(index)
    elif label == "smurf.":
      smurf_indices.append(index)
    elif label == "neptune.":
      neptune_indices.append(index)
    else:
      other_indices.append(index)
    index = index + 1

  #normal_indices = random.sample(normal_indices, 100000)
  #smurf_indices = random.sample(smurf_indices, 100000)
  #neptune_indices = random.sample(neptune_indices, 100000)
  samples = normal_indices + smurf_indices + neptune_indices + other_indices

  dataset = []
  for sample in samples:
    dataset.append(data[sample])
  dict_1, dict_2, dict_3 = get_dicts(dataset)

  for sample in dataset:
    sample[1] = dict_1[sample[1]]
    sample[2] = dict_2[sample[2]]
    sample[3] = dict_3[sample[3]]

  N = len(dataset)
  training_count, validation_count, test_count = dataset_count(N, 0.6, 0.2)
  training_indices, validation_indices, test_indices = partition_data(range(N), training_count, validation_count, test_count)

  samples, labels = [], []
  for sample in dataset:
    data = []
    for d in sample[:-1]:
      data.append(str(d))
    samples.append(','.join(data))
    labels.append(get_label(sample[41]))
  create_dataset(samples, labels, training_indices, validation_indices, test_indices)

if __name__ == "__main__":
  main()