Transformer-Based Model for Malicious URL Classification
In recent years, cyber threats including malicious software, virus, spam, and phishing have grown aggressively via compromised Uniform Resource Locators (URLs). However, the current phishing URL detection solutions based on supervised learning use labeled data for training and classification, leadin...
Saved in:
| Main Authors: | , , , , |
|---|---|
| Other Authors: | |
| Format: | Conference Paper |
| Published: |
Institute of Electrical and Electronics Engineers Inc.
2024
|
| Subjects: | |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| id |
my.uniten.dspace-34380 |
|---|---|
| record_format |
dspace |
| spelling |
my.uniten.dspace-343802024-10-14T11:19:24Z Transformer-Based Model for Malicious URL Classification Do N.Q. Selamat A. Lim K.C. Krejcar O. Ghani N.A.M. 57283917100 24468984100 57889660500 14719632500 57215593148 malicious URL classification natural language processing phishing detection transformer model unsupervised learning Classification (of information) Computer crime Learning algorithms Learning systems Natural language processing systems Supervised learning Viruses Zero-day attack 'current Cyber threats Labeled data Language processing Malicious uniform resource locator classification Natural language processing Natural languages Phishing Phishing detections Transformer modeling Deep learning In recent years, cyber threats including malicious software, virus, spam, and phishing have grown aggressively via compromised Uniform Resource Locators (URLs). However, the current phishing URL detection solutions based on supervised learning use labeled data for training and classification, leading to the dependency on known attacking patterns. These approaches have limitations in fighting against evolving phishing tactics, resulting in a lack of robustness and sustainability. In this study, an unsupervised transformer model is proposed to address the drawbacks of the existing methods which use supervised learning to combat zero-day phishing attacks. Specifically, Bidirectional Encoder Representations from Transformers (BERT) is adopted in this paper to classify malicious URLs. The proposed model was trained on a public dataset and benchmarked with various baseline models using several performance metrics. Results obtained from the experiments showed that BERT-Medium achieved the highest detection accuracy of 98.55% among numerous transformer based models and outperformed other text embedding and deep learning techniques, indicating that the proposed solution is effective and robust in detecting phishing URLs. � 2023 IEEE. Final 2024-10-14T03:19:24Z 2024-10-14T03:19:24Z 2023 Conference Paper 10.1109/ICOCO59262.2023.10397705 2-s2.0-85184851119 https://www.scopus.com/inward/record.uri?eid=2-s2.0-85184851119&doi=10.1109%2fICOCO59262.2023.10397705&partnerID=40&md5=e95b171838ae85d7e157717e8b8fd6f6 https://irepository.uniten.edu.my/handle/123456789/34380 323 327 Institute of Electrical and Electronics Engineers Inc. Scopus |
| institution |
Universiti Tenaga Nasional |
| building |
UNITEN Library |
| collection |
Institutional Repository |
| continent |
Asia |
| country |
Malaysia |
| content_provider |
Universiti Tenaga Nasional |
| content_source |
UNITEN Institutional Repository |
| url_provider |
http://dspace.uniten.edu.my/ |
| topic |
malicious URL classification natural language processing phishing detection transformer model unsupervised learning Classification (of information) Computer crime Learning algorithms Learning systems Natural language processing systems Supervised learning Viruses Zero-day attack 'current Cyber threats Labeled data Language processing Malicious uniform resource locator classification Natural language processing Natural languages Phishing Phishing detections Transformer modeling Deep learning |
| spellingShingle |
malicious URL classification natural language processing phishing detection transformer model unsupervised learning Classification (of information) Computer crime Learning algorithms Learning systems Natural language processing systems Supervised learning Viruses Zero-day attack 'current Cyber threats Labeled data Language processing Malicious uniform resource locator classification Natural language processing Natural languages Phishing Phishing detections Transformer modeling Deep learning Do N.Q. Selamat A. Lim K.C. Krejcar O. Ghani N.A.M. Transformer-Based Model for Malicious URL Classification |
| description |
In recent years, cyber threats including malicious software, virus, spam, and phishing have grown aggressively via compromised Uniform Resource Locators (URLs). However, the current phishing URL detection solutions based on supervised learning use labeled data for training and classification, leading to the dependency on known attacking patterns. These approaches have limitations in fighting against evolving phishing tactics, resulting in a lack of robustness and sustainability. In this study, an unsupervised transformer model is proposed to address the drawbacks of the existing methods which use supervised learning to combat zero-day phishing attacks. Specifically, Bidirectional Encoder Representations from Transformers (BERT) is adopted in this paper to classify malicious URLs. The proposed model was trained on a public dataset and benchmarked with various baseline models using several performance metrics. Results obtained from the experiments showed that BERT-Medium achieved the highest detection accuracy of 98.55% among numerous transformer based models and outperformed other text embedding and deep learning techniques, indicating that the proposed solution is effective and robust in detecting phishing URLs. � 2023 IEEE. |
| author2 |
57283917100 |
| author_facet |
57283917100 Do N.Q. Selamat A. Lim K.C. Krejcar O. Ghani N.A.M. |
| format |
Conference Paper |
| author |
Do N.Q. Selamat A. Lim K.C. Krejcar O. Ghani N.A.M. |
| author_sort |
Do N.Q. |
| title |
Transformer-Based Model for Malicious URL Classification |
| title_short |
Transformer-Based Model for Malicious URL Classification |
| title_full |
Transformer-Based Model for Malicious URL Classification |
| title_fullStr |
Transformer-Based Model for Malicious URL Classification |
| title_full_unstemmed |
Transformer-Based Model for Malicious URL Classification |
| title_sort |
transformer-based model for malicious url classification |
| publisher |
Institute of Electrical and Electronics Engineers Inc. |
| publishDate |
2024 |
| _version_ |
1814061118923997184 |
| score |
13.211869 |