An attack detection framework based on BERT and deep learning

Basit Öğe Kaydı
dc.authorid0000-0002-6490-0396
dc.contributor.authorSeyyar, Yunus Emre
dc.contributor.authorYavuz, Ali Gökhan
dc.contributor.authorÜnver, Halil Murat
dc.date.accessioned2022-11-15T08:05:45Z
dc.date.available2022-11-15T08:05:45Z
dc.date.issued2022
dc.departmentTAÜ, Mühendislik Fakültesi, Bilgisayar Mühendisliği Bölümüen_US
dc.description.abstractDeep Learning (DL) and Natural Language Processing (NLP) techniques are improving and enriching with a rapid pace. Furthermore, we witness that the use of web applications is increasing in almost every direction in parallel with the related technologies. Web applications encompass a wide array of use cases utilizing personal, financial, defense, and political information (e.g., wikileaks incident). Indeed, to access and to manipulate such information are among the primary goals of attackers. Thus, vulnerability of the information targeted by adversaries is a vital problem and if such information is captured then the consequences can be devastating, which can, potentially, become national security risks in the extreme cases. In this study, as a remedy to this problem, we propose a novel model that is capable of distinguishing normal HTTP requests and anomalous HTTP requests. Our model employs NLP techniques, Bidirectional Encoder Representations from Transformers (BERT) model, and DL techniques. Our experimental results reveal that the proposed approach achieves a success rate over 99.98% and an F1 score over 98.70% in the classification of anomalous and normal requests. Furthermore, web attack detection time of our model is significantly lower (i.e., 0.4 ms) than the other approaches presented in the literature.
dc.identifier.citationDeep Learning (DL) and Natural Language Processing (NLP) techniques are improving and enriching with a rapid pace. Furthermore, we witness that the use of web applications is increasing in almost every direction in parallel with the related technologies. Web applications encompass a wide array of use cases utilizing personal, financial, defense, and political information (e.g., wikileaks incident). Indeed, to access and to manipulate such information are among the primary goals of attackers. Thus, vulnerability of the information targeted by adversaries is a vital problem and if such information is captured then the consequences can be devastating, which can, potentially, become national security risks in the extreme cases. In this study, as a remedy to this problem, we propose a novel model that is capable of distinguishing normal HTTP requests and anomalous HTTP requests. Our model employs NLP techniques, Bidirectional Encoder Representations from Transformers (BERT) model, and DL techniques. Our experimental results reveal that the proposed approach achieves a success rate over 99.98% and an F1 score over 98.70% in the classification of anomalous and normal requests. Furthermore, web attack detection time of our model is significantly lower (i.e., 0.4 ms) than the other approaches presented in the literature.
dc.identifier.doi10.1109/ACCESS.2022.3185748
dc.identifier.scopus2-s2.0-85133648584
dc.identifier.scopusqualityN/A
dc.identifier.urihttps://hdl.handle.net/20.500.12846/686
dc.identifier.volume10en_US
dc.identifier.wosWOS:000838524100001
dc.identifier.wosqualityQ2
dc.indekslendigikaynakWeb of Science
dc.indekslendigikaynakScopus
dc.institutionauthorYavuz, Ali Gökhan
dc.language.isoen
dc.publisherIEEE
dc.relation.ispartofIEEE Access
dc.relation.publicationcategoryMakale - Uluslararası Hakemli Dergi - Kurum Öğretim Elemanı
dc.rightsinfo:eu-repo/semantics/openAccess
dc.subjectBit Error Rateen_US
dc.subjectNatural Language Processingen_US
dc.subjectUniform Resource Locatorsen_US
dc.subjectStructured Query Languageen_US
dc.subjectBitfehlerrateen_US
dc.subjectVerarbeitung Natürlicher Spracheen_US
dc.subjectEinheitliche Ressourcenlokatorenen_US
dc.subjectStrukturierte Abfragespracheen_US
dc.subjectBit Hata Oranıen_US
dc.subjectDoğal Dil İşlemeen_US
dc.subjectTekdüzen Kaynak Konum Belirleyicilerien_US
dc.subjectYapılandırılmış Sorgu Dilien_US
dc.titleAn attack detection framework based on BERT and deep learning
dc.typeArticle

Dosyalar

Orijinal paket
Listeleniyor 1 - 1 / 1
Küçük Resim
İsim:
An_Attack_Detection_Framework_Based_on_BERT_and_Deep_Learning.pdf
Boyut:
2.1 MB
Biçim:
Adobe Portable Document Format
Açıklama:
Tam Metin / Full Text
İndir
Lisans paketi
Listeleniyor 1 - 1 / 1
[ X ]
İsim:
license.txt
Boyut:
1.44 KB
Biçim:
Item-specific license agreed upon to submission
Açıklama:
İndir

Koleksiyon

WoS İndeksli Yayınlar Koleksiyonu
Bilgisayar Mühendisliği Bölümü Koleksiyonu
Scopus İndeksli Yayınlar Koleksiyonu