Phishing Link Detection Using Ensemble Learning

The conflict of having to become safer on the internet is one. continually expanding, and phishing attacks are a huge part of the concern. As for the current work, several aspects of supervised learning are explored in the present paper. Decision Tree and others are generally convenient for detecting phishing links. RF has been developed in computer science along with Support Vector Machine (SVM), Naive Bayes, k-nearest Neighbors (KNN), and linear support vector classifier. (LSVC). The validity of each model is then discussed as well as their strength and possible weaknesses in identifying These factors are used especially, when targets of phishing attempts are highlighted. So to improve the results further, we recommend an ensemble of three classifiers – Random Forest, Logistic Regression, and Gradient Boost are the top three models. It is an ensemble technique that uses predictions of several base estimators to enhance the sort of generality or stability compared to another estima to or, for a model compared to the others or compared to a multivariate The study also reveals that ensemble design strategies enhance the performance of the systems. and as such enhance the current phishing detection systems with better defenses against such attacks. Therefore, the results of this research will improve and advance. measures related to security and greatly stress the advisability of using various methods existing and in use in the contemporary currents of cyber security.