Author Information

individual

What campus are you from?

Daytona Beach

Authors' Class Standing

Justin Villa, Senior

Lead Presenter's Name

Justin Villa

Faculty Mentor Name

Laxima Niure Kandel

Abstract

Phishing attacks have evolved into a sophisticated cybersecurity threat, targeting individuals and organizations through emails, websites and social media platforms. Current detection methods rely on predefined rules and deep learning models, but they struggle to identify new and evolving phishing techniques. This research proposes an adaptive hybrid AI approach that integrates Random Forest (RF) and Long Short-Term Memory (LSTM) neural networks. RF is effective for analyzing structured phishing data, while LSTM can detect patterns in sequential data, such as phishing messages. To improve accuracy and adaptability, we will incorporate adversarial training and explainable AI techniques. A key contribution of this research is the creation of a continuously updated phishing dataset, which integrates website URLs and social media phishing messages. By addressing the limitations of static datasets, our model will be better equipped to detect new phishing threats in real time. This project aims to enhance phishing detection capabilities, improve AI transparency, and provide a practical security solution for online users.

Did this research project receive funding support from the Office of Undergraduate Research.

No

Share

COinS
 

Harnessing Hybrid-AI for Real-Time Phishing Detection

Phishing attacks have evolved into a sophisticated cybersecurity threat, targeting individuals and organizations through emails, websites and social media platforms. Current detection methods rely on predefined rules and deep learning models, but they struggle to identify new and evolving phishing techniques. This research proposes an adaptive hybrid AI approach that integrates Random Forest (RF) and Long Short-Term Memory (LSTM) neural networks. RF is effective for analyzing structured phishing data, while LSTM can detect patterns in sequential data, such as phishing messages. To improve accuracy and adaptability, we will incorporate adversarial training and explainable AI techniques. A key contribution of this research is the creation of a continuously updated phishing dataset, which integrates website URLs and social media phishing messages. By addressing the limitations of static datasets, our model will be better equipped to detect new phishing threats in real time. This project aims to enhance phishing detection capabilities, improve AI transparency, and provide a practical security solution for online users.

 

To view the content in your browser, please download Adobe Reader or, alternately,
you may Download the file to your hard drive.

NOTE: The latest versions of Adobe Reader do not support viewing PDF files within Firefox on Mac OS and if you are using a modern (Intel) Mac, there is no official plugin for viewing PDF files within the browser window.