AntiPhishX: An AI-driven service-oriented ensemble framework for detecting phishing and ai-powered phishing attacks

IF 4.3 2区计算机科学 Q2 COMPUTER SCIENCE, INFORMATION SYSTEMS

Information and Software Technology Pub Date : 2025-08-27 DOI:10.1016/j.infsof.2025.107877

Abdul Malik , Bilal Khan , Saeed Mian Qaisar , Moez Krichen

{"title":"AntiPhishX: An AI-driven service-oriented ensemble framework for detecting phishing and ai-powered phishing attacks","authors":"Abdul Malik , Bilal Khan , Saeed Mian Qaisar , Moez Krichen","doi":"10.1016/j.infsof.2025.107877","DOIUrl":null,"url":null,"abstract":"<div><h3>Context</h3><div>The internet has become an essential societal utility, providing opportunities for both legitimate and illegitimate users. Cyberattacks, including phishing Uniform Resource Locator (URL) attacks, have emerged as a significant cybersecurity concern, especially with the increasing adoption of Artificial Intelligence (AI). The exponential growth of AI-driven phishing URL attacks presents new challenges for cyberspace security.</div></div><div><h3>Objective</h3><div>This study aims to develop a novel approach, named AntiPhishX, to detect phishing and AI-phishing URL attacks effectively. The model leverages advancements in AI and service-oriented computing to enhance detection accuracy and overcome the limitations of existing methods.</div></div><div><h3>Methods</h3><div>The proposed AntiPhishX approach integrates Natural Language Processing (NLP) techniques to extract relevant features and analyze text dependencies within URLs. A cohesive model is designed by applying machine learning (ML) algorithms to the processed feature sets. A voting-based ensemble of best-performing ML models is constructed to classify URLs as phishing, AI-phishing, or benign in real time. The model is implemented and evaluated in Python using a dataset of 90,000 URLs collected from the PhishTank platform.</div></div><div><h3>Results</h3><div>The AntiPhishX model outperformed benchmark models, achieving: Precision: 98.32 %, Recall: 97.63 %, F-score: 98.31 %, and Detection rate: 98.12 %</div></div><div><h3>Conclusion</h3><div>The findings demonstrate the potential of AI-driven and service-oriented computing approaches, such as AntiPhishX, in strengthening cyberspace defenses against evolving phishing threats. This study highlights the effectiveness of integrating NLP and ML techniques in phishing URL detection systems.</div></div>","PeriodicalId":54983,"journal":{"name":"Information and Software Technology","volume":"188 ","pages":"Article 107877"},"PeriodicalIF":4.3000,"publicationDate":"2025-08-27","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"0","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"Information and Software Technology","FirstCategoryId":"94","ListUrlMain":"https://www.sciencedirect.com/science/article/pii/S0950584925002162","RegionNum":2,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"Q2","JCRName":"COMPUTER SCIENCE, INFORMATION SYSTEMS","Score":null,"Total":0}

引用次数: 0

Abstract

Context

The internet has become an essential societal utility, providing opportunities for both legitimate and illegitimate users. Cyberattacks, including phishing Uniform Resource Locator (URL) attacks, have emerged as a significant cybersecurity concern, especially with the increasing adoption of Artificial Intelligence (AI). The exponential growth of AI-driven phishing URL attacks presents new challenges for cyberspace security.

Objective

This study aims to develop a novel approach, named AntiPhishX, to detect phishing and AI-phishing URL attacks effectively. The model leverages advancements in AI and service-oriented computing to enhance detection accuracy and overcome the limitations of existing methods.

Methods

The proposed AntiPhishX approach integrates Natural Language Processing (NLP) techniques to extract relevant features and analyze text dependencies within URLs. A cohesive model is designed by applying machine learning (ML) algorithms to the processed feature sets. A voting-based ensemble of best-performing ML models is constructed to classify URLs as phishing, AI-phishing, or benign in real time. The model is implemented and evaluated in Python using a dataset of 90,000 URLs collected from the PhishTank platform.

Results

The AntiPhishX model outperformed benchmark models, achieving: Precision: 98.32 %, Recall: 97.63 %, F-score: 98.31 %, and Detection rate: 98.12 %

Conclusion

The findings demonstrate the potential of AI-driven and service-oriented computing approaches, such as AntiPhishX, in strengthening cyberspace defenses against evolving phishing threats. This study highlights the effectiveness of integrating NLP and ML techniques in phishing URL detection systems.

查看原文本刊更多论文

AntiPhishX：一个ai驱动的面向服务的集成框架，用于检测网络钓鱼和ai驱动的网络钓鱼攻击

互联网已经成为一种重要的社会工具，为合法和非法用户提供了机会。网络攻击，包括网络钓鱼统一资源定位器（URL）攻击，已经成为一个重要的网络安全问题，特别是随着人工智能（AI）的日益普及。人工智能驱动的网络钓鱼URL攻击呈指数级增长，对网络空间安全提出了新的挑战。本研究旨在开发一种名为AntiPhishX的新方法，以有效检测网络钓鱼和人工智能网络钓鱼URL攻击。该模型利用人工智能和面向服务的计算的进步来提高检测精度并克服现有方法的局限性。方法提出的AntiPhishX方法集成了自然语言处理（NLP）技术，提取相关特征并分析url中的文本依赖关系。通过将机器学习算法应用于处理后的特征集，设计了内聚模型。构建了一个基于投票的最佳ML模型集合，用于实时将url分类为网络钓鱼、人工智能网络钓鱼或良性。该模型使用从PhishTank平台收集的90,000个url的数据集在Python中实现和评估。结果AntiPhishX模型优于基准模型，准确率：98.32%，召回率：97.63%,F-score: 98.31%，检出率：98.12%。结论研究结果表明，人工智能驱动和面向服务的计算方法，如AntiPhishX，在加强网络空间防御不断发展的网络钓鱼威胁方面具有潜力。本研究强调了在网络钓鱼URL检测系统中整合自然语言处理和机器学习技术的有效性。

本文章由计算机程序翻译，如有差异，请以英文原文为准。

求助全文

约1分钟内获得全文求助全文

来源期刊

Information and Software Technology 工程技术-计算机：软件工程

CiteScore

9.10

自引率

7.70%

发文量

164

审稿时长

9.6 weeks

期刊介绍： Information and Software Technology is the international archival journal focusing on research and experience that contributes to the improvement of software development practices. The journal''s scope includes methods and techniques to better engineer software and manage its development. Articles submitted for review should have a clear component of software engineering or address ways to improve the engineering and management of software development. Areas covered by the journal include: • Software management, quality and metrics, • Software processes, • Software architecture, modelling, specification, design and programming • Functional and non-functional software requirements • Software testing and verification & validation • Empirical studies of all aspects of engineering and managing software development Short Communications is a new section dedicated to short papers addressing new ideas, controversial opinions, "Negative" results and much more. Read the Guide for authors for more information. The journal encourages and welcomes submissions of systematic literature studies (reviews and maps) within the scope of the journal. Information and Software Technology is the premiere outlet for systematic literature studies in software engineering.