def get_mail_files():
for i in range(1, 46):
try:
# mail = mailparser.parse_from_file('../../Mailboxes/PhishingCorpus_Jose_Nazario/public_phishing/phishing3/{}.eml'.format(i))
mail = mailparser.parse_from_file('../../Mailboxes/IndividualTestMails/Phish/{}.eml'.format(i))
# mail = mailparser.parse_from_file('../../Mailboxes/IndividualTestMails/Ham/{}.eml'.format(i))
# mail = mailparser.parse_from_file('../../Mailboxes/Jonathan_Mailbox/{}.eml'.format(i))
if 'ARC-Authentication-Results' in mail.headers or 'Authentication-Results' in mail.headers:
try:
headers = mail.headers['ARC-Authentication-Results']
except KeyError:
headers = mail.headers['Authentication-Results']
else:
headers = None
test_mail_item = EmailData( \
mail.subject, \
mail.from_, \
mail.attachments, \
mail.body, \
headers
)
test_mail_item.generate_features()
except FileNotFoundError:
pass
def test_model_olden_phish(model, test_data_dir, start, end):
count = 0
phish = 0
for i in range(start, end+1):
try:
mail = mailparser.parse_from_file(r'{}{}.eml'.format(test_data_dir, i))
test_mail = EmailData(mail.subject, mail.from_, mail.attachments, mail.body, mail.headers)
test_mail.generate_features()
result = model.predict(test_mail.repr_in_arr())
count+=1
if result == 1:
phish+=1
print("Result: {}".format(result))
except FileNotFoundError:
pass
print("Detected Mails: {} -- Total Mails: {}".format(phish, count))
print("Accuracy: {}".format((phish/count)*100))
def main():
train_email_data = EmailData()
train_email_data.load_from_file('data/train')
feature = Feature()
feature.learn(train_email_data)
train_data_set = feature.translate_email_data(train_email_data)
#print(feature.features)
naive_bayesian = NaiveBayesian()
naive_bayesian.learn(feature, train_data_set)
test_email_data = EmailData()
test_email_data.load_from_file('data/test')
test_data_set = feature.translate_email_data(test_email_data)
print('# Training set')
test(naive_bayesian, train_data_set)
print('# Testing set')
test(naive_bayesian, test_data_set)
from EmailData import EmailData
import re
import mailparser
for i in range(1, 57):
try:
# mail = mailparser.parse_from_file('../../Mailboxes/PhishingCorpus_Jose_Nazario/public_phishing/phishing3/{}.eml'.format(i))
# mail = mailparser.parse_from_file('../../Mailboxes/PhishingCorpus_Jose_Nazario/public_phishing/phishing3/{}.eml'.format(i))
mail = mailparser.parse_from_file(
'../../Mailboxes/Phish/ModernPhish3/{}.eml'.format(i))
# mail = mailparser.parse_from_file('../../Mailboxes/Hams/ModernHam1/{}.eml'.format(i))
# mail = mailparser.parse_from_file('../../Mailboxes/enron_mail_20150507/maildir/allen-p/all_documents/{}..eml'.format(i))
# mail = mailparser.parse_from_file('../../Mailboxes/enron_mail_20150507/maildir/arnold-j/all_documents/{}..eml'.format(i))
test_mail_item = EmailData( \
mail.subject, \
mail.from_, \
mail.attachments, \
mail.body, \
mail.headers
)
test_mail_item.generate_features()
print("{}".format(test_mail_item))
except FileNotFoundError:
pass
def test_model_single(model: RandomForestClassifier, file_path: str):
mail = mailparser.parse_from_file(r'{}'.format(file_path))
test_mail = EmailData(mail.subject, mail.from_, mail.attachments, mail.body, mail.headers)
test_mail.generate_features()
result = model.predict(test_mail.repr_in_arr())
print("Result: {}".format(result))