Protection Against Spear Phishing Attacks Using the Ensemble Method of Machine Learning - MSc Project Submission Sheet

Sep 17, 2026

Uploaded by Theodora

This study investigates and implements ensemble machine learning methods to enhance the detection and classification of spear phishing emails, addressing a critical cybersecurity challenge driven by increasingly sophisticated attacks. The research aims to improve feature extraction, deploy ensemble learning models, and deliver a practical detection system. Text is converted into numerical vectors using TF-IDF, and LightGBM is identified as the most effective model, outperforming Logistic Regression and Naive Bayes in experiments. A Gradio-based interface enables real-time classification for practical organizational deployment, while future work will explore adaptive learning to improve resilience against evolving phishing tactics.

LanguageEnglish

CategoryResearch & Report

FormatPDF

Pages36

Reading Time1 h 31 min

AccessFree to download

Abstract

1 Introduction

Phishing and attack context

Types and categories of phishing

Protection Against Spear Phishing Attacks Using the Ensemble Method of Machine Learning Jecinta Ifechukwu Fidelis Student ID: 23148306 School of Computing National College of Ireland Supervisor:Imran Khan National College of Ireland MSc Project Submission Sheet School of Computing Jecinta Ifechukwu Fidelis Student Name: 23148306 Student ID: ……………………………………………………………………………………………………………..…… Msc in Cyber Security 2024 Programme: ……………………………………………………………… Year: ………………………….. Msc Research Practicum Module: Imran Khan Lecturer: Submission 16/09/2024 Due Date: ………………………………………………………………………………………………………….……… Project Title: ………………………………………………………………………………………………………….……… Word Count: ……………………………………… Page Count: ………………………………….…….……… I hereby certify that the information contained in this (my submission) is information pertaining to research I conducted for this project. All information other than my own contribution will be fully referenced and listed in the relevant bibliography section at the rear of the project. ALL internet material must be referenced in the bibliography section. Students are required to use the Referencing Standard specified in the report template. To use other author's written or electronic work is illegal (plagiarism) and may result in disciplinary action. Jecinta Ifechukwu Fidelis Signature: 16/09/2024 Date: PLEASE READ THE FOLLOWING INSTRUCTIONS AND CHECKLIST | Attach a completed copy of this sheet to each project (including multiple copies) | □ | | --- | --- | | Attach a Moodle submission receipt of the online project submission, to each project (including multiple copies) . | □ | | You must ensure that you retain a HARD COPY of the project, both for your own reference and in case a project is lost or mislaid. It is not sufficient to keep a copy on computer. | □ | Assignments that are submitted to the Programme Coordinator Office must be placed into the assignment box located outside the office. | Office Use Only | | | --- | --- | | Signature: | | | Date: | | | Penalty Applied (if applicable): | | Protection Against Spear Phishing Attacks Using the Ensemble Method of Machine Learning Jecinta Ifechukwu Fidelis 23148306 Abstract This study investigated and implemented the use of multiple ensemble machine learning methods to enhance the detection and classification of spear phishing emails, which is a critical challenge in cybersecurity due to the sophisticated nature of phishing attacks. The primary motivation for this study is the need for more effective phishing detection systems that can accurately distinguish between legitimate emails and spear phishing attempts beyond the human eyes. To address this issue, the study focused on the following three key objectives: enhancing feature extraction techniques, implementing ensemble machine learning models, and deploying a practical phishing detection system. The research employed advanced text feature extraction methods with specific regard to Term Frequency-Inverse Document Frequency (TF-IDF), to convert email content into numerical vectors, thereby improving model accuracy. In this study, LightGBM emerged as the most effective model in test experiments and outperforming traditional models like Logistic Regression and Naive Bayes. To ensure practical applicability, this model was deployed with a user-friendly interface developed using Gradio, enabling real-time email classification. This integration provides a practical solution for organisations to combat spear phishing attacks. The study's findings demonstrate the efficacy of ensemble models in improving phishing detection and offer significant implications for both academic research and practical applications. Future work will explore adaptive learning approaches to further enhance the system's resilience against evolving phishing tactics and address limitations identified in dynamic environments. Key words: Ensemble Machine L

FAQs

What problem does the project address?

The project targets the difficulty of detecting and classifying spear phishing emails, which are designed to appear legitimate and trick recipients into revealing sensitive information.

How is email content represented for the models?

The study converts email text into numerical vectors using TF-IDF feature extraction.

Which model performs best and how is it deployed?

LightGBM performs best in the reported experiments. It is deployed with a user-friendly Gradio interface for real-time email classification.

Sep 17, 2026

Uploaded by Theodora