Institutional Repository of Vilnius University
Comparison of phishing url recognition methods /
Abstract
dc:descriptionA phishing attack is a social engineering cyberattack designed to exfiltrate users’ confidential information, such as login credentials, financial information, or personal data. Phishing attacks are some of the most common cybersecurity threats. As phishing methods evolve, phishing detection methods must improve in order to keep up with the threat. This work evaluates the phishing website URL detection capabilities of various machine learning models. Specifically, the work focuses on comparing the capabilities of traditional machine learning algorithms and modern large language models. The experiments compare 7 well-known machine learning algorithms and 6 pre-trained large language models from the Hugging Face platform. Results show that some large language models are highly capable in detecting phishing URLs, even without specialised training. However, traditional methods still offer competitive performance at much lower computational cost.
Degree
thesis:*- Grantor dc:publisher
- Institutional Repository of Vilnius University
- Year dc:date
- 2026
Author and committee
dc:creator, dc:contributor.*- Author dc:creator
-
- Žukauskas, Laurynas,
Subjects
dc:subject × 1Rights
dc:rights- Statement dc:rights
-
- info:eu-repo/semantics/openAccess
- Language dc:language
- eng
Identifiers
dc:identifier.*- Repository record dc:identifier
- https://repository.vu.lt/VU:ELABAETD308339406&prefLang=en_US