Abstract:
Day after day the number of internet users increases, phishing has grown increasingly
breakneck. Phishing attacks pose a serious threat to people’s daily lives and the online
environment. For example, the attacker poses as a trustworthy source in order to get
sensitive information or the victim’s digital identity, such as a credit card number or
certificate or other valuable information. For this reason, people lose their identity after
falling into the trap of these raiders. As the name implies, phishing or faking sites are false
copies of actual web sites. When a person’s identification card gets stolen, they are
cheating. To create the website for this paper debate publishing, we will be relying on a
machine learning algorithm, Neural Network Classifier MLPC (Multilayer perceptron
Classifier) and have differentiated the percentage of accuracy between them. We have
used five machine learning algorithms: Naive Bayes algorithm, K-nearest neighbors
(KNN), SVM, Decision tree, Random forest algorithm. Most accurate and well directed
perspective of this approach may be found in our dataset that it’s a scam or fake website.
Among them, the Random Forest algorithm provided 97.9 % accuracy.
