Bidirectional LSTM: An Innovative Approach for Phishing URL Identification

Baojiang Cui,Jiangdong Deng,Yuchen Liang

doi:10.1007/978-3-030-22263-5_32

Bidirectional LSTM: An Innovative Approach for Phishing URL Identification

Baojiang Cui, Jiangdong Deng + Show 1 more

https://doi.org/10.1007/978-3-030-22263-5_32

Copy DOI

Publication Date: Jun 19, 2019

Citations: 10

Affiliation: Beijing University of Posts and Telecommunications, Shadyside Hospital

#Bidirectional Long Short-Term Memory Network #Bidirectional Long Short-Term Memory + Show 8 more

Abstract
Full-Text PDF
Similar Papers

Abstract

Today, prevailing phishing scams have caused serious network security problems. By adopting a domain generation algorithm, the cyber attackers are able to disguise their malware in various forms, making it easier to steal personal identity and property. This paper proposed an innovative approach to detect malicious URLs by employing a Bidirectional Long Short-Term Memory (B-LSTM) network. The paper also compares the performance of the B-LSTM network with that of conventional machine learning methods and unidirectional LSTM. The results demonstrate that the B-LSTM network has superior performance than the other machine learning methods, producing a F1 score of 95.9%.

Full Text