Weighted transformer neural network for web attack detection using request URL.
Saved in:
| Title: | Weighted transformer neural network for web attack detection using request URL. |
|---|---|
| Authors: | Deshpande, Kirti V.1 (AUTHOR) kirtivdeshpande@gmail.com, Singh, Jaibir1 (AUTHOR) |
| Source: | Multimedia Tools & Applications. May2024, Vol. 83 Issue 15, p43983-44007. 25p. |
| Subjects: | Transformer models, Uniform Resource Locators, Deep learning, Web-based user interfaces, Mathematical optimization, Generative adversarial networks |
| Abstract: | Web application firewalls (WAFs) and other Intrusion Detection Systems (IDS) techniques are employed to defend the network against web attacks. Even so, attacks may succeed since most WAFs demand extensive configuration expertise that depends on filters. Despite notable successes, deep information has been utilized in varied applications. Still, it's crucial to have a reliable method for detecting the attack due to the attacker's various ways of concealment of the URLs. Several methods were introduced for detecting the attacks in web applications; still, the accuracy of detection and the computation burden are challenging aspects. Hence, a web attack detection mechanism is introduced in this research using the deep learning framework using the URL request. The proposed method utilizes a three-fold attack detection strategy to detect the attack with minimal computation complexity. Initially, the profile is checked to determine the genuinity of a user, and then, the bot scanners are identified using the generalized adversarial network (GAN). Finally, the attack detection is employed using the transformer neural network, wherein the adjustable parameters are modified using the weighted mean of vectors (INFO) optimization technique. The performance of a proposed method is evaluated based on various assessment measures like Accuracy, Precision, Recall, F-Measure, TPR, FPR, FNR and TNR and acquired the values of 99.97%, 99.96%, 99.97%, 99.97%, 99.97%, 0.03%, 0.03%, and 99.97% respectively. [ABSTRACT FROM AUTHOR] |
| Copyright of Multimedia Tools & Applications is the property of Springer Nature and its content may not be copied or emailed to multiple sites without the copyright holder's express written permission. Additionally, content may not be used with any artificial intelligence tools or machine learning technologies. However, users may print, download, or email articles for individual use. This abstract may be abridged. No warranty is given about the accuracy of the copy. Users should refer to the original published version of the material for the full abstract. (Copyright applies to all Abstracts.) | |
| Database: | Engineering Source |
|
Full text is not displayed to guests.
Login for full access.
|
|
| Abstract: | Web application firewalls (WAFs) and other Intrusion Detection Systems (IDS) techniques are employed to defend the network against web attacks. Even so, attacks may succeed since most WAFs demand extensive configuration expertise that depends on filters. Despite notable successes, deep information has been utilized in varied applications. Still, it's crucial to have a reliable method for detecting the attack due to the attacker's various ways of concealment of the URLs. Several methods were introduced for detecting the attacks in web applications; still, the accuracy of detection and the computation burden are challenging aspects. Hence, a web attack detection mechanism is introduced in this research using the deep learning framework using the URL request. The proposed method utilizes a three-fold attack detection strategy to detect the attack with minimal computation complexity. Initially, the profile is checked to determine the genuinity of a user, and then, the bot scanners are identified using the generalized adversarial network (GAN). Finally, the attack detection is employed using the transformer neural network, wherein the adjustable parameters are modified using the weighted mean of vectors (INFO) optimization technique. The performance of a proposed method is evaluated based on various assessment measures like Accuracy, Precision, Recall, F-Measure, TPR, FPR, FNR and TNR and acquired the values of 99.97%, 99.96%, 99.97%, 99.97%, 99.97%, 0.03%, 0.03%, and 99.97% respectively. [ABSTRACT FROM AUTHOR] |
|---|---|
| ISSN: | 13807501 |
| DOI: | 10.1007/s11042-023-17356-9 |
