Malicious URL Detection Using Machine Learning Algorithms

Fuente: Zenodo
Enregistré dans:
Détails bibliographiques
Auteurs principaux: Ashish Nanotkar, Kaustubh Duke, Ketan Pisalkar, Sushil Bhakne, Aditya Wadewale
Format: Recurso digital
Publié: Zenodo 2026
Accès en ligne:
Tags: Ajouter un tag
Pas de tags, Soyez le premier à ajouter un tag!
_version_ 1866901548885344256
author Ashish Nanotkar
Kaustubh Duke
Ketan Pisalkar
Sushil Bhakne
Aditya Wadewale
author_facet Ashish Nanotkar
Kaustubh Duke
Ketan Pisalkar
Sushil Bhakne
Aditya Wadewale
contents <p>The expansive adoption of the internet has led to a surge in cyber risks, such as phishing, malware dissemination, and online scams, with malicious URLs frequently serving as the primary delivery mechanism for these attacks. Traditional blacklist systems are often inadequate because they cannot keep pace with the constant generation and modification of harmful links. This research introduces an efficient, machine learning-driven framework for the identification of malicious URLs. The method presented focuses on analyzing the inherent lexical and structural properties of a URL, avoiding the need to inspect webpage content, thereby significantly speeding up the detection process. The system employs multiple machine learning models, trained on these characteristics, to categorize URLs as either safe or dangerous. Engineered for real-time operation, the framework is capable of processing vast quantities of URL data. Evaluation experiments confirm that the proposed technique delivers strong accuracy alongside a minimal rate of false positives. This system can be integrated into web browsers, email defense platforms, and network security infrastructure to bolster overall digital safety.</p>
format Recurso digital
id zenodo_https___doi_org_10_5281_zenodo_18507798
institution Zenodo
language
publishDate 2026
publisher Zenodo
record_format zenodo
spellingShingle Malicious URL Detection Using Machine Learning Algorithms
Ashish Nanotkar
Kaustubh Duke
Ketan Pisalkar
Sushil Bhakne
Aditya Wadewale
<p>The expansive adoption of the internet has led to a surge in cyber risks, such as phishing, malware dissemination, and online scams, with malicious URLs frequently serving as the primary delivery mechanism for these attacks. Traditional blacklist systems are often inadequate because they cannot keep pace with the constant generation and modification of harmful links. This research introduces an efficient, machine learning-driven framework for the identification of malicious URLs. The method presented focuses on analyzing the inherent lexical and structural properties of a URL, avoiding the need to inspect webpage content, thereby significantly speeding up the detection process. The system employs multiple machine learning models, trained on these characteristics, to categorize URLs as either safe or dangerous. Engineered for real-time operation, the framework is capable of processing vast quantities of URL data. Evaluation experiments confirm that the proposed technique delivers strong accuracy alongside a minimal rate of false positives. This system can be integrated into web browsers, email defense platforms, and network security infrastructure to bolster overall digital safety.</p>
title Malicious URL Detection Using Machine Learning Algorithms
url https://doi.org/10.5281/zenodo.18507798