article · Scientific Reports
This research develops PhishNet 1.0, an integrated cybersecurity framework designed to improve the detection of fraudulent phishing web addresses. The system relies on feature selection using the Boruta method followed by evaluation of multiple machine learning classifiers, identifying Gradient Boosting, Nearest Neighbours, and Decision Trees as the strongest performers. These algorithms were merged into ensemble models, including soft voting and stacking configurations with Logistic Regression serving as the final estimator. To maximise performance, several metaheuristic optimization algorithms were tested to tune the stacking model hyperparameters. The Optuna-optimised stacking classifier delivered the strongest overall outcomes on phishing URL datasets, reaching an accuracy of 96.15 percent alongside balanced precision and recall. The combined approach demonstrates that pairing ensemble classifiers with targeted optimization algorithms substantially enhances automated detection capabilities for deceptive websites.
Phishing web addresses represent a persistent cyber threat used to deceive users and compromise secure information. By combining multiple learning models and fine-tuning them with advanced optimization techniques, this framework reliably spots malicious links with high accuracy. This systematic approach helps protect online users by offering a more robust foundation for automated threat defence.
The framework is designed for integration into real-world cybersecurity systems, targeting security software vendors, enterprise network defenders, and anti-fraud teams. By establishing high accuracy on URL datasets, it appears to be at an applied and tested stage. However, deployment into production security environments would require adapting the pipeline to live data streams and varied network settings.
AI-generated from the published abstract. Always read the original work before citing.
The objective of this research is to enhance phishing detection through ensemble learning integrated with well-structured metaheuristic algorithms. Various classifiers, including Logistic Regression, Nearest Neighbors, Support Vector Machine, Decision Tree, Naïve Bayes, and Gradient Boosting, were evaluated using features selected via the Boruta method. Among these, Gradient Boosting, KNN, and Decision Tree achieved the highest performance. These models were subsequently incorporated into two ensemble classifiers, namely Soft Voting and Stacking, with Logistic Regression selected as the final estimator in the stacking model, which outperformed alternative estimators. Furthermore, several metaheuristic optimization algorithms, such as genetic algorithm (GA), ant colony optimization (ACO), particle swarm optimization (PSO), bayesian optimization, and optuna, were employed to optimize the hyperparameters of the stacking model, thereby improving classification performance. Among these, the optuna-optimized stacking classifier achieved the best results, with an accuracy of 96.15%, precision of 96.45%, recall of 96.68%, and F1 score of 96.56%. This work makes a novel contribution by presenting an integrated framework, PhishNet 1.0, which combines ensemble learning with diverse metaheuristic optimizers and demonstrates its empirical effectiveness on phishing URL datasets. Its practical and experimental focus differentiates it from prior research and establishes a reproducible benchmark for cybersecurity applications. The findings conclusively demonstrate that ensemble learning combined with metaheuristic optimization substantially enhances phishing webpage detection and provides a reliable approach for real-world cybersecurity systems.
This page summarises published work. The authoritative version sits with the publisher.
DOI: 10.1038/s41598-025-31447-7
Is something wrong with this record? Report it or request removal.
Discussion
Have you built on this work, tried to replicate it, or seen it applied in practice? Share what you know. Verified researchers and MARATTO™ domain experts can open a discussion, and any member can reply. Contributions are reviewed before they appear.
No discussion yet. Open the first thread.
New to MARATTO™? Create a free account.
