International Research Journal of Engineering and Technology (IRJET)
e-ISSN: 2395-0056
Volume: 11 Issue: 09 | Sep 2024
p-ISSN: 2395-0072
www.irjet.net
PHISHING WEBSITE DETECTION USING MACHINE LEARNING AND DEEP NEURAL NETWORKS Deepak Kumar Jha1, Pallavi Mishra2, Aradhya A Rathore3, Sankalp Verma4 1-4Vellore Institute of Technology, Vellore, Tamil Nadu, India
---------------------------------------------------------------------***---------------------------------------------------------------------
Abstract -
quarter of 2014, up 10.7% from the previous quarter. 99.4% of phishing sites used port 80, and more than 55% used the name of the target. 123,972 phishing assaults were reported in the second half of 2014; as a result, 4.5 billion dollars in 2014 and 4.6 billion dollars in 2015 were lost financially.
In the digital world, phishing is still a prevalent and developing problem that can be extremely costly for both individuals and businesses. The ever-evolving strategies used by fraudsters make it more difficult to identify phishing assaults with high precision. As a reaction, we suggest creating an intricate phishing detection system that makes use of cutting-edge machine learning techniques. Our method looks for trends and abnormalities in large datasets to increase the efficacy and accuracy of phishing attempt detection. Our solution aims to offer a comprehensive defence mechanism that can adapt to the dynamic nature of phishing attacks, protecting users and organisations from sophisticated and newly emerging phishing schemes as digital environments get more complicated.
The following image shows a simplified version of how phishing websites operate.
Key Words: Phishing Detection, Machine Learning, Cybersecurity, Fraud Prevention, Digital Threats, Attack Identification, Data Analysis, Threat Mitigation 1. INTRODUCTION 1.1 Brief History Fig-1: Illustration of a Phishing Attack Workflow
Early in the 1990s, America Online (AOL) computers were the target of the first known phishing assault [37]. By establishing fictitious accounts with erroneous credit card information, attackers took advantage of AOL's initial validation procedure. These accounts were used to access AOL resources after they were activated. In response, AOL strengthened its verification processes; however, hackers adjusted by obtaining personal data from actual users. They obtained user credentials by impersonating AOL staff in phishing emails and instant chats. This strategy spread to encompass a number of e-commerce and banking websites.
2. METHODOLOGY 2.1 Workflow
1.2 Statistics 2.97 billion people, or more than 38% of the world's population, were online as of 2014. These users have been the subject of phishing schemes more often, which have resulted in large financial losses. Phishing attacks increased 160% in 2012 over 2011. Approximately 450,000 phishing assaults resulted in losses over $5.9 billion in 2013. There were 125,215 attacks in the first
© 2024, IRJET
|
Impact Factor value: 8.226
Fig-2: Flowchart of the Phishing Detection System
|
ISO 9001:2008 Certified Journal
|
Page 614