Title: Anti-phishing model based on relative content mining
Authors: Parvinder Singh; Bhawna Sharma
Addresses: Department of Computer Science and Engineering, DeenbandhuChhotu Ram University of Science and Technology, Murthal, Sonepat, Haryana, India ' Department of Computer Science and Engineering, DeenbandhuChhotu Ram University of Science and Technology, Murthal, Sonepat, Haryana, India
Abstract: Phishing has attracted larger section of researchers and application developers not due to the rising instances of phishing attacks but also due to the sophisticated techniques that are being employed to execute on the attack. To address one of the diverse mechanisms of phishing attacks, the authors have proposed an anti-phishing model for detecting phishing URLs using relative content mining. The relative similarity calculation method uses a combination of cosine similarity and Jaccard similarity. Machine learning oriented feed forward back propagation neural networks (FFBPNN) in combination with support vector machine (SVM) algorithms are used as an anti-phishing technique. A hybrid training and classification algorithm using three kernels namely linear, polynomial and radial basis function (RBF) are implemented. The proposed approach provides best solution for the detection of the phisher in the cyber world. Multiple scenarios such as precision and accuracy are calculated to evaluate the proposed work. Precision of the proposed work is 0.781456 for the detection of cyber-attacks.
Keywords: phishing; machine learning; training; classification.
International Journal of Computational Vision and Robotics, 2022 Vol.12 No.1, pp.66 - 84
Received: 20 Sep 2020
Accepted: 14 Oct 2020
Published online: 30 Nov 2021 *