def check(url, clf):
# url=sys.argv[1]
fT = time.time()
features_test = features_extraction.main(url)
feT = time.time()
features_test = np.array(features_test).reshape((1, -1))
prT = time.time()
pred = clf.predict(features_test)
# prob=clf.predict_proba(features_test)
preT = time.time()
# pred=[1]
print("\n")
print("Feature extraction Time = " + str(feT - fT))
print("Prediction Time = " + str(preT - prT))
print("Total Time = " + str(preT - fT))
print("\n")
if int(pred[0]) == 1:
print(url)
print("This is a safe website.")
return False
# return str(url)+" is a safe website."
elif int(pred[0]) == -1:
print(url)
print("This is a phishing website..!")
return True
def check_website(request):
"""
"""
# import ipdb; ipdb.set_trace()
url = request.GET.get('url')
#url='https://www.google.com/'
features_test = features_extraction.main(url)
# Due to updates to scikit-learn, we now need a 2D array as a parameter to the predict function.
features_test = np.array(features_test).reshape((1, -1))
# import ipdb; ipdb.set_trace()
clf = joblib.load(os.getcwd() + '/backend/classifier/random_forest.pkl')
pred = clf.predict(features_test)
# Print the probability of prediction (if needed)
# prob = clf.predict_proba(features_test)
# print 'Features=', features_test, 'The predicted probability is - ', prob, 'The predicted label is - ', pred
# print "The probability of this site being a phishing website is ", features_test[0]*100, "%"
if int(pred[0]) == 1:
# print "The website is safe to browse"
print("SAFE")
return HttpResponse("SAFE")
elif int(pred[0]) == -1:
print(" PHISING")
# print "The website has phishing features. DO NOT VISIT!"
return HttpResponse('Phising')
def main(url):
#url = sys.argv[1]
features_test = features_extraction.main(url)
# Due to updates to scikit-learn, we now need a 2D array as a parameter to the predict function.
features_test = np.array(features_test).reshape((1, -1))
clf = joblib.load('classifier/random_forest.pkl')
pred = clf.predict(features_test)
# Print the probability of prediction (if needed)
# prob = clf.predict_proba(features_test)
# print 'Features=', features_test, 'The predicted probability is - ', prob, 'The predicted label is - ', pred
# print "The probability of this site being a phishing website is ", features_test[0]*100, "%"
if int(pred[0]) == 1:
# print "The website is safe to browse"
print("SAFE")
return "SAFE"
elif int(pred[0]) == -1:
# print "The website has phishing features. DO NOT VISIT!"
print("PHISHING")
return "MALICIOUS"
def get_prediction_from_url(test_url):
features_test = features_extraction.main(test_url)
# Due to updates to scikit-learn, we now need a 2D array as a parameter to the predict function.
features_test = np.array(features_test).reshape((1, -1))
clf = joblib.load(LOCALHOST_PATH + DIRECTORY_NAME +
'/classifier/random_forest.pkl')
print(clf)
def get_prediction_from_url(test_url,html):
features_test = features_extraction.main(test_url,html)
features_test = np.array(features_test).reshape((1, -1))
#localPath = os.getcwd()
clf = joblib.load(WORKING_PATH + '/classifier/random_forest1.pkl')
pred = clf.predict(features_test)
return int(pred[0])
def get_prediction_from_url(test_url):
features_test = features_extraction.main(test_url)
features_test = np.array(features_test).reshape((1, -1))
clf = joblib.load(LOCALHOST_PATH + DIRECTORY_NAME +
'/classifier/random_forest.pkl')
pred = clf.predict(features_test)
return int(pred[0])
def get_prediction_from_url(test_url):
features_test = features_extraction.main(test_url)
# we now need a 2D array as a parameter to the predict function.
features_test = np.array(features_test).reshape((1, -1))
clf = joblib.load(LOCALHOST_PATH + '/classifier/finalized_model.sav')
pred = clf.predict(features_test)
return int(pred[0])
def predict(test_url):
features_test = features_extraction.main(test_url)
features_test = np.array(features_test).reshape((1, -1))
clf = joblib.load(MODEL_PATH)
pred = clf.predict(features_test)
print(test_url)
print(pred)
return int(pred[0])
def main():
url = "http://www.spit.ac.in"
features_test = features_extraction.main(url)
clf = joblib.load('classifier/extratree.pkl')
features_test = [features_test]
pred = clf.predict(features_test)
prob = clf.predict_proba(features_test)
# print 'Features=', features_test, 'The predicted probability is - ', prob, 'The predicted label is - ', pred
# print "The probability of this site being a phishing website is ", features_test[0]*100, "%"
if int(pred[0]) == 1:
# print "The website is safe to browse"
print("SAFE")
elif int(pred[0]) == -1:
# print "The website has phishing features. DO NOT VISIT!"
print("PHISHING")
def main():
url = sys.argv[1]
features_test = features_extraction.main(url)
clf = joblib.load('classifier/random_forest.pkl')
pred = clf.predict(features_test)
prob = clf.predict_proba(features_test)
# print 'Features=', features_test, 'The predicted probability is - ', prob, 'The predicted label is - ', pred
# print "The probability of this site being a phishing website is ", features_test[0]*100, "%"
if int(pred[0]) == 1:
# print "The website is safe to browse"
print "SAFE"
elif int(pred[0]) == -1:
# print "The website has phishing features. DO NOT VISIT!"
print "PHISHING"
def main():
url = sys.argv[1]
features_test = features_extraction.main(url)
# 2d array per scikit-learn
features_test = np.array(features_test).reshape((1, -1))
clf = joblib.load(LOCALHOST_PATH + DIRECTORY_NAME +
'/classifier/random_forest_new_4.pkl')
prediction = clf.predict(features_test)
prediction_int = int(prediction[0])
probability = clf.predict_proba(features_test)
respond_data = {
"features": features_test.tolist(),
"probability": probability.tolist(),
"prediction": prediction_int,
"url": url
}
json_respond_data = json.dumps(respond_data)
print(json_respond_data)
def main():
url = sys.argv[1]
#url='https://www.google.com/'
features_test = features_extraction.main(url)
# Due to updates to scikit-learn, we now need a 2D array as a parameter to the predict function.
features_test = np.array(features_test).reshape((1, -1))
clf = joblib.load(
'C://xampp//htdocs//webc3//classifier//random_forest.pkl')
pred = clf.predict(features_test)
# Print the probability of prediction (if needed)
# prob = clf.predict_proba(features_test)
# print 'Features=', features_test, 'The predicted probability is - ', prob, 'The predicted label is - ', pred
# print "The probability of this site being a phishing website is ", features_test[0]*100, "%"
if int(pred[0]) == 1:
# print "The website is safe to browse"
print("SAFE")
elif int(pred[0]) == -1:
# print "The website has phishing features. DO NOT VISIT!"
print("PHishing")
