Location
Accra, Ghana and Virtual
Start Date
28-8-2025 4:15 PM
End Date
28-8-2025 4:45 PM
Description
Phishing attacks pose significant risks to cybersecurity, exploiting user trust through deceptive email content. This paper presents a machine learning based framework for detecting phishing emails using a 2024 dataset comprising over 80,000 labeled samples sourced from PhishTank and Kaggle. Features were engineered from URLs, email content, and metadata. Five models— Logistic Regression, Support Vector Machine (SVM), Random Forest, XGBoost, and K-Nearest Neighbors (KNN)—were evaluated. Simulated results demonstrate that ensemble models, particularly Random Forest and XGBoost, delivered optimal results, with near-perfect accuracy and recall. The study highlights the efficacy of combining feature-based engineering with ensemble learning to enhance real-time phishing detection.
Machine Learning Driven Email Phishing Detection
Accra, Ghana and Virtual
Phishing attacks pose significant risks to cybersecurity, exploiting user trust through deceptive email content. This paper presents a machine learning based framework for detecting phishing emails using a 2024 dataset comprising over 80,000 labeled samples sourced from PhishTank and Kaggle. Features were engineered from URLs, email content, and metadata. Five models— Logistic Regression, Support Vector Machine (SVM), Random Forest, XGBoost, and K-Nearest Neighbors (KNN)—were evaluated. Simulated results demonstrate that ensemble models, particularly Random Forest and XGBoost, delivered optimal results, with near-perfect accuracy and recall. The study highlights the efficacy of combining feature-based engineering with ensemble learning to enhance real-time phishing detection.
