Treffer: SafeLink AI: URL Threat Detection using LLM

Title:
SafeLink AI: URL Threat Detection using LLM
Authors:
Vardhelli, Lokesh Goud, Vennu, Vaishnavi, Guthikonda, Neeraja, Pagidimarri, Aswanth Narayana
Contributors:
Vardhelli, Lokesh Goud, Vennu, Vaishnavi, Guthikonda, Neeraja, Pagidimarri, Aswanth Narayana, N., Lavanya
Publisher Information:
Zenodo
Publication Year:
2025
Collection:
Zenodo
Subject Terms:
Cybersecurity, machine learning, URL detection, phishing, natural language processing, transformer models, SafeLink AI
Document Type:
Report report
Language:
English
Relation:
https://zenodo.org/records/15366518; oai:zenodo.org:15366518; https://doi.org/10.5281/zenodo.15366518
DOI:
10.5281/zenodo.15366518
Availability:
https://doi.org/10.5281/zenodo.15366518
https://zenodo.org/records/15366518
Rights:
Creative Commons Attribution 4.0 International ; cc-by-4.0 ; https://creativecommons.org/licenses/by/4.0/legalcode ; © 2025 Vardhelli Lokesh Goud, Vennu Vaishnavi, Guthikonda Neeraja, Pagidimarri Aswanth Narayana
Accession Number:
edsbas.3A5003B9
Database:
BASE

Weitere Informationen

SafeLink AI is a real-time URL threat detection system that utilizes machine learning and large language models (LLMs) to accurately classify malicious and benign URLs. This B.Tech final year project integrates NLP techniques, WHOIS data analysis, and a user-friendly interface to provide instant and intelligent cybersecurity threat assessments. Achieving over 95% classification accuracy, the project is implemented with a Python Flask backend and features future scalability for browser integration. This report documents the design, development, evaluation, and deployment aspects of SafeLink AI.