Lightweight URL Phishing Detection System Using SVM and Similarity

Computer Engineering

As a criminal offense of employing technical method to thieve sensitive information of customers, phishing is presently an important risk facing the net, and losses because of phishing are developing regularly. function engineering is essential in phishing website detection answers; however, the accuracy of detection critically relies upon on earlier knowledge of features. furthermore, even though features extracted for one-of-a-kind dimensions are more comprehensive, a disadvantage is that extracting those functions calls for a massive amount of time. To address those barriers, we propose a multidimensional characteristic phishing detection method based on a quick detection approach by way of the usage of deep mastering (MFPD). within the first step, character sequence functions of the given URL are extracted and used for brief class by means of deep mastering, and this step does not require 1/3-party assistance or any prior know-how approximately phishing. inside the 2d step, we combine URL statistical functions, webpage code features, website text capabilities and the short classification end result of deep studying into multi-dimensional features. The technique can lessen the detection time for putting a threshold

This work is licensed under a Creative Commons Attribution-ShareAlike 4.0 International License.