A proof of concept implementation of Explainable Artificial Intelligence (XAI) in digital forensics

Stuart W. Hall; Amin Sakzad; Sepehr Minagar

doi:10.1007/978-3-031-23020-2_4

A proof of concept implementation of Explainable Artificial Intelligence (XAI) in digital forensics

Stuart W. Hall, Amin Sakzad, Sepehr Minagar

Department of Software Systems & Cybersecurity

Research output: Chapter in Book/Report/Conference proceeding › Conference Paper › Research › peer-review

5 Citations (Scopus)

Abstract

Explainable Artificial Intelligence (XAI) has been a subject of much research related to its potential to have a transformative impact on industries where AI implementations require a high degree of trust. Digital Forensics is one such field, with practitioners dealing with an ever-increasing volume of data and a variety of devices. This research first examines the concept of XAI and what it could accomplish in the Digital Forensics field. Using of series of 23 manufactured virtual hard disk (VHD) images, AI models were trained to make predictions on image and video file content and file system metadata. The manufactured VHD images can be of independent interest for educational purposes as well. The predictions were then processed using the Local Interpretable Model-Agnostic Explanations (LIME) toolkit. These results were then used as the basis for a discussion on the future of XAI in the digital forensics industry and potential areas for further research.

Original language	English
Title of host publication	Network and System Security - 16th International Conference, NSS 2022 Denarau Island, Fiji, December 9–12, 2022 Proceedings
Editors	Xingliang Yuan, Guangdong Bai, Cristina Alcaraz, Suryadipta Majumdar
Place of Publication	Cham Switzerland
Publisher	Springer
Pages	66-85
Number of pages	20
ISBN (Electronic)	9783031230202
ISBN (Print)	9783031230196
DOIs	https://doi.org/10.1007/978-3-031-23020-2_4
Publication status	Published - 2022
Event	International Conference on Network and System Security 2022 - Denarau Island, Fiji Duration: 9 Dec 2022 → 12 Dec 2022 Conference number: 16th https://link.springer.com/book/10.1007/978-3-031-23020-2 (Proceedings) http://nsclab.org/nss2022/ (Website)

Publication series

Name	Lecture Notes in Computer Science
Publisher	Springer
Volume	13787
ISSN (Print)	0302-9743
ISSN (Electronic)	1611-3349

Conference

Conference	International Conference on Network and System Security 2022
Abbreviated title	NSS 2022
Country/Territory	Fiji
City	Denarau Island
Period	9/12/22 → 12/12/22
Internet address	https://link.springer.com/book/10.1007/978-3-031-23020-2 (Proceedings) http://nsclab.org/nss2022/ (Website)

Keywords

Digital forensics
LIME
XAI

Access to Document

10.1007/978-3-031-23020-2_4

Cite this

Hall, S. W., Sakzad, A., & Minagar, S. (2022). A proof of concept implementation of Explainable Artificial Intelligence (XAI) in digital forensics. In X. Yuan, G. Bai, C. Alcaraz, & S. Majumdar (Eds.), Network and System Security - 16th International Conference, NSS 2022 Denarau Island, Fiji, December 9–12, 2022 Proceedings (pp. 66-85). (Lecture Notes in Computer Science; Vol. 13787). Springer. https://doi.org/10.1007/978-3-031-23020-2_4

Hall, Stuart W. ; Sakzad, Amin ; Minagar, Sepehr. / A proof of concept implementation of Explainable Artificial Intelligence (XAI) in digital forensics. Network and System Security - 16th International Conference, NSS 2022 Denarau Island, Fiji, December 9–12, 2022 Proceedings. editor / Xingliang Yuan ; Guangdong Bai ; Cristina Alcaraz ; Suryadipta Majumdar. Cham Switzerland : Springer, 2022. pp. 66-85 (Lecture Notes in Computer Science).

@inproceedings{eec70d567c514a39a88c2dcddf623549,

title = "A proof of concept implementation of Explainable Artificial Intelligence (XAI) in digital forensics",

abstract = "Explainable Artificial Intelligence (XAI) has been a subject of much research related to its potential to have a transformative impact on industries where AI implementations require a high degree of trust. Digital Forensics is one such field, with practitioners dealing with an ever-increasing volume of data and a variety of devices. This research first examines the concept of XAI and what it could accomplish in the Digital Forensics field. Using of series of 23 manufactured virtual hard disk (VHD) images, AI models were trained to make predictions on image and video file content and file system metadata. The manufactured VHD images can be of independent interest for educational purposes as well. The predictions were then processed using the Local Interpretable Model-Agnostic Explanations (LIME) toolkit. These results were then used as the basis for a discussion on the future of XAI in the digital forensics industry and potential areas for further research.",

keywords = "Digital forensics, LIME, XAI",

author = "Hall, {Stuart W.} and Amin Sakzad and Sepehr Minagar",

note = "Publisher Copyright: {\textcopyright} 2022, The Author(s), under exclusive license to Springer Nature Switzerland AG.; International Conference on Network and System Security 2022, NSS 2022 ; Conference date: 09-12-2022 Through 12-12-2022",

year = "2022",

doi = "10.1007/978-3-031-23020-2_4",

language = "English",

isbn = "9783031230196",

series = "Lecture Notes in Computer Science",

publisher = "Springer",

pages = "66--85",

editor = "Xingliang Yuan and Guangdong Bai and Cristina Alcaraz and Suryadipta Majumdar",

booktitle = "Network and System Security - 16th International Conference, NSS 2022 Denarau Island, Fiji, December 9–12, 2022 Proceedings",

url = "https://link.springer.com/book/10.1007/978-3-031-23020-2, http://nsclab.org/nss2022/",

}

Hall, SW, Sakzad, A & Minagar, S 2022, A proof of concept implementation of Explainable Artificial Intelligence (XAI) in digital forensics. in X Yuan, G Bai, C Alcaraz & S Majumdar (eds), Network and System Security - 16th International Conference, NSS 2022 Denarau Island, Fiji, December 9–12, 2022 Proceedings. Lecture Notes in Computer Science, vol. 13787, Springer, Cham Switzerland, pp. 66-85, International Conference on Network and System Security 2022, Denarau Island, Fiji, 9/12/22. https://doi.org/10.1007/978-3-031-23020-2_4

A proof of concept implementation of Explainable Artificial Intelligence (XAI) in digital forensics. / Hall, Stuart W.; Sakzad, Amin; Minagar, Sepehr.
Network and System Security - 16th International Conference, NSS 2022 Denarau Island, Fiji, December 9–12, 2022 Proceedings. ed. / Xingliang Yuan; Guangdong Bai; Cristina Alcaraz; Suryadipta Majumdar. Cham Switzerland: Springer, 2022. p. 66-85 (Lecture Notes in Computer Science; Vol. 13787).

Research output: Chapter in Book/Report/Conference proceeding › Conference Paper › Research › peer-review

TY - GEN

T1 - A proof of concept implementation of Explainable Artificial Intelligence (XAI) in digital forensics

AU - Hall, Stuart W.

AU - Sakzad, Amin

AU - Minagar, Sepehr

N1 - Conference code: 16th

PY - 2022

Y1 - 2022

N2 - Explainable Artificial Intelligence (XAI) has been a subject of much research related to its potential to have a transformative impact on industries where AI implementations require a high degree of trust. Digital Forensics is one such field, with practitioners dealing with an ever-increasing volume of data and a variety of devices. This research first examines the concept of XAI and what it could accomplish in the Digital Forensics field. Using of series of 23 manufactured virtual hard disk (VHD) images, AI models were trained to make predictions on image and video file content and file system metadata. The manufactured VHD images can be of independent interest for educational purposes as well. The predictions were then processed using the Local Interpretable Model-Agnostic Explanations (LIME) toolkit. These results were then used as the basis for a discussion on the future of XAI in the digital forensics industry and potential areas for further research.

AB - Explainable Artificial Intelligence (XAI) has been a subject of much research related to its potential to have a transformative impact on industries where AI implementations require a high degree of trust. Digital Forensics is one such field, with practitioners dealing with an ever-increasing volume of data and a variety of devices. This research first examines the concept of XAI and what it could accomplish in the Digital Forensics field. Using of series of 23 manufactured virtual hard disk (VHD) images, AI models were trained to make predictions on image and video file content and file system metadata. The manufactured VHD images can be of independent interest for educational purposes as well. The predictions were then processed using the Local Interpretable Model-Agnostic Explanations (LIME) toolkit. These results were then used as the basis for a discussion on the future of XAI in the digital forensics industry and potential areas for further research.

KW - Digital forensics

KW - LIME

KW - XAI

UR - http://www.scopus.com/inward/record.url?scp=85145020097&partnerID=8YFLogxK

U2 - 10.1007/978-3-031-23020-2_4

DO - 10.1007/978-3-031-23020-2_4

M3 - Conference Paper

AN - SCOPUS:85145020097

SN - 9783031230196

T3 - Lecture Notes in Computer Science

SP - 66

EP - 85

BT - Network and System Security - 16th International Conference, NSS 2022 Denarau Island, Fiji, December 9–12, 2022 Proceedings

A2 - Yuan, Xingliang

A2 - Bai, Guangdong

A2 - Alcaraz, Cristina

A2 - Majumdar, Suryadipta

PB - Springer

CY - Cham Switzerland

T2 - International Conference on Network and System Security 2022

Y2 - 9 December 2022 through 12 December 2022

ER -

Hall SW, Sakzad A, Minagar S. A proof of concept implementation of Explainable Artificial Intelligence (XAI) in digital forensics. In Yuan X, Bai G, Alcaraz C, Majumdar S, editors, Network and System Security - 16th International Conference, NSS 2022 Denarau Island, Fiji, December 9–12, 2022 Proceedings. Cham Switzerland: Springer. 2022. p. 66-85. (Lecture Notes in Computer Science). doi: 10.1007/978-3-031-23020-2_4