Intelligent Predictive Analytics Model for Detecting and Preventing Phishing Attacks in Institutional Networks (Published)
Phishing attacks remain one of the most persistent and damaging cybersecurity threats affecting institutional networks worldwide. With the increasing sophistication of social engineering techniques and malicious web infrastructures, traditional rule-based and signature-based detection systems have become insufficient. This study proposes an intelligent predictive analytics model for detecting and preventing phishing attacks within institutional environments. The model leverages supervised machine learning techniques to analyze URL- and content-based features for accurate phishing classification. A dataset containing 2,200 labeled instances was used, and key features were selected through preprocessing and dimensionality reduction techniques. Two supervised learning models; Random Forest (RF) and Support Vector Machine (SVM) were implemented and evaluated using standard performance metrics including accuracy, precision, recall, and F1-score. Experimental results demonstrate that the RF model outperformed SVM, achieving an accuracy of 95.7% compared to 93.3% for SVM. The findings confirm that intelligent predictive analytics significantly enhances phishing detection accuracy and provides a scalable, adaptive solution for institutional cybersecurity systems.
Keywords: institutional networks, intelligent predictive analytics, model detecting, preventing phishing attacks