An Integrated, Smart & Dynamic Prediction of Phishing Website using LSTM

  • R. Padmanabhan, Suvetha Suresh, Siddharth Swaminathan, Pavan Balaji K.

Abstract

With burgeoning technological advancements and rapid development of web applications, internet users utilize these benefits to a great extent. This increases the chance of the user to get caught in the web of phishing. Phishing is one of the most well-known cybercrime activities where people are misled into the wrong sites by using fraudulent methods. The aim of these websites is to impound the user’s personal information and financial details for personal benefits or misuse.  Several machine-learning techniques and approaches have been proposed to detect phishing. These techniques depend on features that are extracted from website samples. However, some of these approaches are effective but the major drawback is that the system is constrained to some features. Sometimes an appropriate feature is not selected to detect phishing. In this work, we investigate the features that should be selected to detect phishing websites by using Fuzzy Logic algorithm and also use Long Short-Term Memory (LSTM) framework to classify the websites. The proposed system, based on a large-scale data set compiled from real phishing cases, has shown that our system can effectively prevent phishing attacks and improve network security.

Published
2020-06-06
How to Cite
R. Padmanabhan, Suvetha Suresh, Siddharth Swaminathan, Pavan Balaji K. (2020). An Integrated, Smart & Dynamic Prediction of Phishing Website using LSTM. International Journal of Advanced Science and Technology, 29(04), 3012 -. Retrieved from http://sersc.org/journals/index.php/IJAST/article/view/24097