Leveraging Federated Learning for File Fragments Classification Based on Depthwise Separable Convolutions

Soha B. Sandouka; Muhamad Felemban

doi:10.1145/3644713.3644807

Leveraging Federated Learning for File Fragments Classification Based on Depthwise Separable Convolutions

Soha B. Sandouka^*, Muhamad Felemban

^*Corresponding author for this work

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

Abstract

In digital forensics, file fragment classification plays a crucial role in the file carving process. Recently, convolutional neural network based models have been utilized for file fragment classification to improve the classification accuracy. However, training CNN models requires a large dataset, presenting a challenge in digital forensics where data is sensitive and confidential. To this end, we propose a federated learning framework for file fragments classification based on depth-wise separable convolutions. Accordingly, we can develop a file fragment classification model that is both privacy-preserving and computationally efficient. We experimentally tested the proposed framework using FFT-75 dataset. The experimental results show that the proposed framework achieves comparable accuracies to those of centralized training models while preserving the privacy and confidentiality of sensitive data.

Original language	English
Title of host publication	ICFNDS 2023 - 2023 The 7th International Conference on Future Networks and Distributed Systems
Publisher	Association for Computing Machinery
Pages	626-632
Number of pages	7
ISBN (Electronic)	9798400709036
DOIs	https://doi.org/10.1145/3644713.3644807
State	Published - 21 Dec 2023
Event	7th International Conference on Future Networks and Distributed Systems, ICFNDS 2023 - Dubai, United Arab Emirates Duration: 21 Dec 2023 → 22 Dec 2023

Publication series

Name	ACM International Conference Proceeding Series

Conference

Conference	7th International Conference on Future Networks and Distributed Systems, ICFNDS 2023
Country/Territory	United Arab Emirates
City	Dubai
Period	21/12/23 → 22/12/23

Bibliographical note

Publisher Copyright:
© 2023 ACM.

Keywords

Deep Learning
Depth-wise Separable Convolution
Digital Forensics
Federated Learning
File Carving
File Fragments Classification

ASJC Scopus subject areas

Human-Computer Interaction
Computer Networks and Communications
Computer Vision and Pattern Recognition
Software

Access to Document

10.1145/3644713.3644807

Cite this

Sandouka, S. B., & Felemban, M. (2023). Leveraging Federated Learning for File Fragments Classification Based on Depthwise Separable Convolutions. In ICFNDS 2023 - 2023 The 7th International Conference on Future Networks and Distributed Systems (pp. 626-632). (ACM International Conference Proceeding Series). Association for Computing Machinery. https://doi.org/10.1145/3644713.3644807

@inproceedings{d086049d51d0441a9f1648606d52b735,

title = "Leveraging Federated Learning for File Fragments Classification Based on Depthwise Separable Convolutions",

abstract = "In digital forensics, file fragment classification plays a crucial role in the file carving process. Recently, convolutional neural network based models have been utilized for file fragment classification to improve the classification accuracy. However, training CNN models requires a large dataset, presenting a challenge in digital forensics where data is sensitive and confidential. To this end, we propose a federated learning framework for file fragments classification based on depth-wise separable convolutions. Accordingly, we can develop a file fragment classification model that is both privacy-preserving and computationally efficient. We experimentally tested the proposed framework using FFT-75 dataset. The experimental results show that the proposed framework achieves comparable accuracies to those of centralized training models while preserving the privacy and confidentiality of sensitive data.",

keywords = "Deep Learning, Depth-wise Separable Convolution, Digital Forensics, Federated Learning, File Carving, File Fragments Classification",

author = "Sandouka, \{Soha B.\} and Muhamad Felemban",

note = "Publisher Copyright: {\textcopyright} 2023 ACM.; 7th International Conference on Future Networks and Distributed Systems, ICFNDS 2023 ; Conference date: 21-12-2023 Through 22-12-2023",

year = "2023",

month = dec,

day = "21",

doi = "10.1145/3644713.3644807",

language = "English",

series = "ACM International Conference Proceeding Series",

publisher = "Association for Computing Machinery",

pages = "626--632",

booktitle = "ICFNDS 2023 - 2023 The 7th International Conference on Future Networks and Distributed Systems",

}

Sandouka, SB & Felemban, M 2023, Leveraging Federated Learning for File Fragments Classification Based on Depthwise Separable Convolutions. in ICFNDS 2023 - 2023 The 7th International Conference on Future Networks and Distributed Systems. ACM International Conference Proceeding Series, Association for Computing Machinery, pp. 626-632, 7th International Conference on Future Networks and Distributed Systems, ICFNDS 2023, Dubai, United Arab Emirates, 21/12/23. https://doi.org/10.1145/3644713.3644807

Leveraging Federated Learning for File Fragments Classification Based on Depthwise Separable Convolutions. / Sandouka, Soha B.; Felemban, Muhamad.
ICFNDS 2023 - 2023 The 7th International Conference on Future Networks and Distributed Systems. Association for Computing Machinery, 2023. p. 626-632 (ACM International Conference Proceeding Series).

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

TY - GEN

T1 - Leveraging Federated Learning for File Fragments Classification Based on Depthwise Separable Convolutions

AU - Sandouka, Soha B.

AU - Felemban, Muhamad

PY - 2023/12/21

Y1 - 2023/12/21

N2 - In digital forensics, file fragment classification plays a crucial role in the file carving process. Recently, convolutional neural network based models have been utilized for file fragment classification to improve the classification accuracy. However, training CNN models requires a large dataset, presenting a challenge in digital forensics where data is sensitive and confidential. To this end, we propose a federated learning framework for file fragments classification based on depth-wise separable convolutions. Accordingly, we can develop a file fragment classification model that is both privacy-preserving and computationally efficient. We experimentally tested the proposed framework using FFT-75 dataset. The experimental results show that the proposed framework achieves comparable accuracies to those of centralized training models while preserving the privacy and confidentiality of sensitive data.

AB - In digital forensics, file fragment classification plays a crucial role in the file carving process. Recently, convolutional neural network based models have been utilized for file fragment classification to improve the classification accuracy. However, training CNN models requires a large dataset, presenting a challenge in digital forensics where data is sensitive and confidential. To this end, we propose a federated learning framework for file fragments classification based on depth-wise separable convolutions. Accordingly, we can develop a file fragment classification model that is both privacy-preserving and computationally efficient. We experimentally tested the proposed framework using FFT-75 dataset. The experimental results show that the proposed framework achieves comparable accuracies to those of centralized training models while preserving the privacy and confidentiality of sensitive data.

KW - Deep Learning

KW - Depth-wise Separable Convolution

KW - Digital Forensics

KW - Federated Learning

KW - File Carving

KW - File Fragments Classification

UR - https://www.scopus.com/pages/publications/85193500495

U2 - 10.1145/3644713.3644807

DO - 10.1145/3644713.3644807

M3 - Conference contribution

AN - SCOPUS:85193500495

T3 - ACM International Conference Proceeding Series

SP - 626

EP - 632

BT - ICFNDS 2023 - 2023 The 7th International Conference on Future Networks and Distributed Systems

PB - Association for Computing Machinery

T2 - 7th International Conference on Future Networks and Distributed Systems, ICFNDS 2023

Y2 - 21 December 2023 through 22 December 2023

ER -

Leveraging Federated Learning for File Fragments Classification Based on Depthwise Separable Convolutions

Abstract

Publication series

Conference

Bibliographical note

Keywords

ASJC Scopus subject areas

Access to Document

Fingerprint

Cite this