Phishing Email Detection Using Improved RCNN Model With Multilevel Vectors and Attention Mechanism

Abstract

The phishing email is one of the significant threats in the world today and has caused tremendous financial losses. Although the methods of confrontation are continually being updated, the results of those methods are not very satisfactory at present. Moreover, phishing emails are growing at an alarming rate in recent years. Therefore, more effective phishing detection technology is needed to curb the threat of phishing emails. In this paper, we first analyzed the email structure. Then, based on an improved recurrent convolutional neural networks (RCNN) model with multilevel vectors and attention mechanism, we proposed a new phishing email detection model named THEMIS, which is used to model emails at the email header, the email body, the character level, and the word level simultaneously. To evaluate the effectiveness of THEMIS, we use an unbalanced dataset that has realistic ratios of phishing and legitimate emails. The experimental results show that the overall accuracy of THEMIS reaches 99.848%. Meanwhile, the false positive rate (FPR) is 0.043%. High accuracy and low FPR ensure that the filter can identify phishing emails with high probability and filter out legitimate emails as little as possible. This promising result is superior to the existing detection methods and verifies the effectiveness of THEMIS in detecting phishing emails.

The Answer is in the Text: Multi-Stage Methods for Phishing Detection Based on Feature Engineering

“…This approach attained an F1-measure of 98.63% and an accuracy rate of 98.91%. Through the use of Recurrent Convolutional Neural Networks (RCNN), Fang et al [46] proposed THEMIS, that employing Word2Vec models e-mails at four levels simultaneously (header, body, character, and word). Its best mark was an F1-Score of 99.31% and an accuracy of 99.84%.…”

A novel hybrid approach of SVM combined with NLP and probabilistic neural network for email phishing

“…Tests were led on two generally utilized supervised spam classifier algorithms: NB & SVM. introduced another phishing email recognition model named THEMIS, which is utilized to show emails at the email header, the email body, the character level & the word level at the same time [20]. Focused on the latest advancement over inquires about concerning machine learning for big data processing and various methods with regards to present day processing situations for different social applications [23].…”

A Dynamic and Combined Phishing Detection Technique

“…The increased use of internet and its applications has increased the opportunities for the cyber attackers to steal users' data and misuse them .Phishing attack has become the most prevalent and successful technique adopted by the cyber criminals to entrap the users .Many predicting mechanisms and algorithms using machine learning have been developed for predicting various types of possible phishing attacks have been proposed by researchers . In this paper [1], an effective mechanism to predict email phishing using deep learning algorithm called THEMIS has been proposed. In this technique, email body and email headers are modeled at character and word level and RCNN is used to predict the phishing email.…”