Data Driven based Malicious URL Detection using Explainable AI

Saranda Poddar; Deepraj Chowdhury; Ashutosh Dhar Dwivedi; Raghava Rao Mukkamala

doi:10.1109/TrustCom56396.2022.00176

Data Driven based Malicious URL Detection using Explainable AI

Saranda Poddar, Deepraj Chowdhury, Ashutosh Dhar Dwivedi^*, Raghava Rao Mukkamala

^*Corresponding author for this work

Research output: Contribution to book/anthology/report/conference proceeding › Article in proceeding › Research › peer-review

4 Citations (Scopus)

Abstract

With the ever-increasing reach of the internet, and its increasing access through various types of devices, the spread of malware, phishing attempts, etc. have steadily been increasing, along with their level of sophistication. Thus it becomes very important to conduct research on different methods to prevent such harmful attacks on systems and users. Using a malicious URL is the common way for hackers to attack a system, thus, to accommodate the variety attack vectors of malicious websites, 21 features were extracted from 651,191 URLs to train the proposed model. A two-stage stacked ensemble learning model, based on gradient boosting methods and random forest, has been trained and tested in the 70:30 ratio of the 651,191 URLs, and an accuracy of 97% has been achieved. Then Explainable AI (XAI) has been used to clearly explain the working of the model, and study the impact of each of the 21 features on the 4 class predictions (benign, defacement, phishing and malware).

Original language	English
Title of host publication	Proceedings - 2022 IEEE 21st International Conference on Trust, Security and Privacy in Computing and Communications, TrustCom 2022
Number of pages	7
Publisher	IEEE Signal Processing Society
Publication date	2022
Pages	1266-1272
ISBN (Electronic)	9781665494250
DOIs	https://doi.org/10.1109/TrustCom56396.2022.00176
Publication status	Published - 2022
Event	21st IEEE International Conference on Trust, Security and Privacy in Computing and Communications, TrustCom 2022 - Virtual, Online, China Duration: 9 Dec 2022 → 11 Dec 2022

Conference

Conference	21st IEEE International Conference on Trust, Security and Privacy in Computing and Communications, TrustCom 2022
Country/Territory	China
City	Virtual, Online
Period	09/12/2022 → 11/12/2022
Sponsor	et al., Huazhong University of Science and Technology, IEEE, IEEE Computer Society, School of Cyber Science and Engineering (CSE), HUST, TCSC IEEE

Series	Proceedings - 2022 IEEE 21st International Conference on Trust, Security and Privacy in Computing and Communications, TrustCom 2022

Bibliographical note

Publisher Copyright:
© 2022 IEEE.

Keywords

ensemble-learning
explainable AI
gradient boosting
malicious URL detection
random forest

Access to Document

10.1109/TrustCom56396.2022.00176

AUB Link

Search for the material in Aalborg University Library's search engine

Cite this

Poddar, Saranda ; Chowdhury, Deepraj ; Dwivedi, Ashutosh Dhar et al. / Data Driven based Malicious URL Detection using Explainable AI. Proceedings - 2022 IEEE 21st International Conference on Trust, Security and Privacy in Computing and Communications, TrustCom 2022. IEEE Signal Processing Society, 2022. pp. 1266-1272 (Proceedings - 2022 IEEE 21st International Conference on Trust, Security and Privacy in Computing and Communications, TrustCom 2022).

@inproceedings{4e54383fc6fa4e3590a7b9c758c3b4c3,

title = "Data Driven based Malicious URL Detection using Explainable AI",

abstract = "With the ever-increasing reach of the internet, and its increasing access through various types of devices, the spread of malware, phishing attempts, etc. have steadily been increasing, along with their level of sophistication. Thus it becomes very important to conduct research on different methods to prevent such harmful attacks on systems and users. Using a malicious URL is the common way for hackers to attack a system, thus, to accommodate the variety attack vectors of malicious websites, 21 features were extracted from 651,191 URLs to train the proposed model. A two-stage stacked ensemble learning model, based on gradient boosting methods and random forest, has been trained and tested in the 70:30 ratio of the 651,191 URLs, and an accuracy of 97% has been achieved. Then Explainable AI (XAI) has been used to clearly explain the working of the model, and study the impact of each of the 21 features on the 4 class predictions (benign, defacement, phishing and malware).",

keywords = "ensemble-learning, explainable AI, gradient boosting, malicious URL detection, random forest",

author = "Saranda Poddar and Deepraj Chowdhury and Dwivedi, {Ashutosh Dhar} and Mukkamala, {Raghava Rao}",

note = "Publisher Copyright: {\textcopyright} 2022 IEEE.; 21st IEEE International Conference on Trust, Security and Privacy in Computing and Communications, TrustCom 2022 ; Conference date: 09-12-2022 Through 11-12-2022",

year = "2022",

doi = "10.1109/TrustCom56396.2022.00176",

language = "English",

series = "Proceedings - 2022 IEEE 21st International Conference on Trust, Security and Privacy in Computing and Communications, TrustCom 2022",

pages = "1266--1272",

booktitle = "Proceedings - 2022 IEEE 21st International Conference on Trust, Security and Privacy in Computing and Communications, TrustCom 2022",

publisher = "IEEE Signal Processing Society",

address = "United States",

}

Poddar, S, Chowdhury, D, Dwivedi, AD & Mukkamala, RR 2022, Data Driven based Malicious URL Detection using Explainable AI. in Proceedings - 2022 IEEE 21st International Conference on Trust, Security and Privacy in Computing and Communications, TrustCom 2022. IEEE Signal Processing Society, Proceedings - 2022 IEEE 21st International Conference on Trust, Security and Privacy in Computing and Communications, TrustCom 2022, pp. 1266-1272, 21st IEEE International Conference on Trust, Security and Privacy in Computing and Communications, TrustCom 2022, Virtual, Online, China, 09/12/2022. https://doi.org/10.1109/TrustCom56396.2022.00176

Data Driven based Malicious URL Detection using Explainable AI. / Poddar, Saranda; Chowdhury, Deepraj; Dwivedi, Ashutosh Dhar et al.
Proceedings - 2022 IEEE 21st International Conference on Trust, Security and Privacy in Computing and Communications, TrustCom 2022. IEEE Signal Processing Society, 2022. p. 1266-1272 (Proceedings - 2022 IEEE 21st International Conference on Trust, Security and Privacy in Computing and Communications, TrustCom 2022).

Research output: Contribution to book/anthology/report/conference proceeding › Article in proceeding › Research › peer-review

TY - GEN

T1 - Data Driven based Malicious URL Detection using Explainable AI

AU - Poddar, Saranda

AU - Chowdhury, Deepraj

AU - Dwivedi, Ashutosh Dhar

AU - Mukkamala, Raghava Rao

PY - 2022

Y1 - 2022

N2 - With the ever-increasing reach of the internet, and its increasing access through various types of devices, the spread of malware, phishing attempts, etc. have steadily been increasing, along with their level of sophistication. Thus it becomes very important to conduct research on different methods to prevent such harmful attacks on systems and users. Using a malicious URL is the common way for hackers to attack a system, thus, to accommodate the variety attack vectors of malicious websites, 21 features were extracted from 651,191 URLs to train the proposed model. A two-stage stacked ensemble learning model, based on gradient boosting methods and random forest, has been trained and tested in the 70:30 ratio of the 651,191 URLs, and an accuracy of 97% has been achieved. Then Explainable AI (XAI) has been used to clearly explain the working of the model, and study the impact of each of the 21 features on the 4 class predictions (benign, defacement, phishing and malware).

AB - With the ever-increasing reach of the internet, and its increasing access through various types of devices, the spread of malware, phishing attempts, etc. have steadily been increasing, along with their level of sophistication. Thus it becomes very important to conduct research on different methods to prevent such harmful attacks on systems and users. Using a malicious URL is the common way for hackers to attack a system, thus, to accommodate the variety attack vectors of malicious websites, 21 features were extracted from 651,191 URLs to train the proposed model. A two-stage stacked ensemble learning model, based on gradient boosting methods and random forest, has been trained and tested in the 70:30 ratio of the 651,191 URLs, and an accuracy of 97% has been achieved. Then Explainable AI (XAI) has been used to clearly explain the working of the model, and study the impact of each of the 21 features on the 4 class predictions (benign, defacement, phishing and malware).

KW - ensemble-learning

KW - explainable AI

KW - gradient boosting

KW - malicious URL detection

KW - random forest

UR - http://www.scopus.com/inward/record.url?scp=85151749362&partnerID=8YFLogxK

U2 - 10.1109/TrustCom56396.2022.00176

DO - 10.1109/TrustCom56396.2022.00176

M3 - Article in proceeding

AN - SCOPUS:85151749362

T3 - Proceedings - 2022 IEEE 21st International Conference on Trust, Security and Privacy in Computing and Communications, TrustCom 2022

SP - 1266

EP - 1272

BT - Proceedings - 2022 IEEE 21st International Conference on Trust, Security and Privacy in Computing and Communications, TrustCom 2022

PB - IEEE Signal Processing Society

T2 - 21st IEEE International Conference on Trust, Security and Privacy in Computing and Communications, TrustCom 2022

Y2 - 9 December 2022 through 11 December 2022

ER -

Poddar S, Chowdhury D, Dwivedi AD, Mukkamala RR. Data Driven based Malicious URL Detection using Explainable AI. In Proceedings - 2022 IEEE 21st International Conference on Trust, Security and Privacy in Computing and Communications, TrustCom 2022. IEEE Signal Processing Society. 2022. p. 1266-1272. (Proceedings - 2022 IEEE 21st International Conference on Trust, Security and Privacy in Computing and Communications, TrustCom 2022). doi: 10.1109/TrustCom56396.2022.00176

Data Driven based Malicious URL Detection using Explainable AI

Abstract

Conference

Bibliographical note

Keywords

Access to Document

AUB Link

Other files and links

Fingerprint

Cite this