Deep Federated Feature Recommendation

Bin Xue; Qinghua Zheng; Zhinan Li; Weihu Zhao; Weihang Zhang; Xue Feng

doi:10.3233/FAIA231248

Deep Federated Feature Recommendation

Bin Xue^*, Qinghua Zheng, Zhinan Li, Weihu Zhao, Weihang Zhang, Xue Feng

^*Corresponding author for this work

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

Abstract

Feature recommendation is one of the most critical and challenging problems in modern digital intelligence system. However, it is difficult to ensure privacy protection in many situations. To address this challenge, a deep federated feature recommendation method, called DF_Rec, is designed. Particularly, deep decentralized federated aggregation learning (DFAL) is jointly developed based on the ingenious combination of several deep frameworks and federated aggregation schemes. Extensive experiments are performed on three authoritative datasets, demonstrating that DF_Rec outperforms existing outstanding systems significantly.

Original language	English
Title of host publication	Electronics, Communications and Networks - Proceedings of the 13th International Conference, CECNet 2023
Editors	Antonio J. Tallon-Ballesteros, Estefania Cortes-Ancos, Diego A. Lopez-Garcia
Publisher	IOS Press BV
Pages	644-649
Number of pages	6
ISBN (Electronic)	9781643684802
DOIs	https://doi.org/10.3233/FAIA231248
Publication status	Published - 12 Jan 2024
Externally published	Yes
Event	13th International Conference on Electronics, Communications and Networks, CECNet 2023 - Hybrid, Macau, China Duration: 17 Nov 2023 → 20 Nov 2023

Publication series

Name	Frontiers in Artificial Intelligence and Applications
Volume	381
ISSN (Print)	0922-6389
ISSN (Electronic)	1879-8314

Conference

Conference	13th International Conference on Electronics, Communications and Networks, CECNet 2023
Country/Territory	China
City	Hybrid, Macau
Period	17/11/23 → 20/11/23

Keywords

deep learning
feature extraction-recommendation
Federated learning

Access to Document

10.3233/FAIA231248

Cite this

Xue, B., Zheng, Q., Li, Z., Zhao, W., Zhang, W., & Feng, X. (2024). Deep Federated Feature Recommendation. In A. J. Tallon-Ballesteros, E. Cortes-Ancos, & D. A. Lopez-Garcia (Eds.), Electronics, Communications and Networks - Proceedings of the 13th International Conference, CECNet 2023 (pp. 644-649). (Frontiers in Artificial Intelligence and Applications; Vol. 381). IOS Press BV. https://doi.org/10.3233/FAIA231248

Xue, Bin ; Zheng, Qinghua ; Li, Zhinan et al. / Deep Federated Feature Recommendation. Electronics, Communications and Networks - Proceedings of the 13th International Conference, CECNet 2023. editor / Antonio J. Tallon-Ballesteros ; Estefania Cortes-Ancos ; Diego A. Lopez-Garcia. IOS Press BV, 2024. pp. 644-649 (Frontiers in Artificial Intelligence and Applications).

@inproceedings{2d891faf918d4009afa54fbf0a8bb2e7,

title = "Deep Federated Feature Recommendation",

abstract = "Feature recommendation is one of the most critical and challenging problems in modern digital intelligence system. However, it is difficult to ensure privacy protection in many situations. To address this challenge, a deep federated feature recommendation method, called DF_Rec, is designed. Particularly, deep decentralized federated aggregation learning (DFAL) is jointly developed based on the ingenious combination of several deep frameworks and federated aggregation schemes. Extensive experiments are performed on three authoritative datasets, demonstrating that DF_Rec outperforms existing outstanding systems significantly.",

keywords = "deep learning, feature extraction-recommendation, Federated learning",

author = "Bin Xue and Qinghua Zheng and Zhinan Li and Weihu Zhao and Weihang Zhang and Xue Feng",

note = "Publisher Copyright: {\textcopyright} 2024 The authors and IOS Press.; 13th International Conference on Electronics, Communications and Networks, CECNet 2023 ; Conference date: 17-11-2023 Through 20-11-2023",

year = "2024",

month = jan,

day = "12",

doi = "10.3233/FAIA231248",

language = "English",

series = "Frontiers in Artificial Intelligence and Applications",

publisher = "IOS Press BV",

pages = "644--649",

editor = "Tallon-Ballesteros, {Antonio J.} and Estefania Cortes-Ancos and Lopez-Garcia, {Diego A.}",

booktitle = "Electronics, Communications and Networks - Proceedings of the 13th International Conference, CECNet 2023",

address = "Netherlands",

}

Xue, B, Zheng, Q, Li, Z, Zhao, W, Zhang, W & Feng, X 2024, Deep Federated Feature Recommendation. in AJ Tallon-Ballesteros, E Cortes-Ancos & DA Lopez-Garcia (eds), Electronics, Communications and Networks - Proceedings of the 13th International Conference, CECNet 2023. Frontiers in Artificial Intelligence and Applications, vol. 381, IOS Press BV, pp. 644-649, 13th International Conference on Electronics, Communications and Networks, CECNet 2023, Hybrid, Macau, China, 17/11/23. https://doi.org/10.3233/FAIA231248

Deep Federated Feature Recommendation. / Xue, Bin; Zheng, Qinghua; Li, Zhinan et al.
Electronics, Communications and Networks - Proceedings of the 13th International Conference, CECNet 2023. ed. / Antonio J. Tallon-Ballesteros; Estefania Cortes-Ancos; Diego A. Lopez-Garcia. IOS Press BV, 2024. p. 644-649 (Frontiers in Artificial Intelligence and Applications; Vol. 381).

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

TY - GEN

T1 - Deep Federated Feature Recommendation

AU - Xue, Bin

AU - Zheng, Qinghua

AU - Li, Zhinan

AU - Zhao, Weihu

AU - Zhang, Weihang

AU - Feng, Xue

PY - 2024/1/12

Y1 - 2024/1/12

N2 - Feature recommendation is one of the most critical and challenging problems in modern digital intelligence system. However, it is difficult to ensure privacy protection in many situations. To address this challenge, a deep federated feature recommendation method, called DF_Rec, is designed. Particularly, deep decentralized federated aggregation learning (DFAL) is jointly developed based on the ingenious combination of several deep frameworks and federated aggregation schemes. Extensive experiments are performed on three authoritative datasets, demonstrating that DF_Rec outperforms existing outstanding systems significantly.

AB - Feature recommendation is one of the most critical and challenging problems in modern digital intelligence system. However, it is difficult to ensure privacy protection in many situations. To address this challenge, a deep federated feature recommendation method, called DF_Rec, is designed. Particularly, deep decentralized federated aggregation learning (DFAL) is jointly developed based on the ingenious combination of several deep frameworks and federated aggregation schemes. Extensive experiments are performed on three authoritative datasets, demonstrating that DF_Rec outperforms existing outstanding systems significantly.

KW - deep learning

KW - feature extraction-recommendation

KW - Federated learning

UR - http://www.scopus.com/inward/record.url?scp=85183076145&partnerID=8YFLogxK

U2 - 10.3233/FAIA231248

DO - 10.3233/FAIA231248

M3 - Conference contribution

AN - SCOPUS:85183076145

T3 - Frontiers in Artificial Intelligence and Applications

SP - 644

EP - 649

BT - Electronics, Communications and Networks - Proceedings of the 13th International Conference, CECNet 2023

A2 - Tallon-Ballesteros, Antonio J.

A2 - Cortes-Ancos, Estefania

A2 - Lopez-Garcia, Diego A.

PB - IOS Press BV

T2 - 13th International Conference on Electronics, Communications and Networks, CECNet 2023

Y2 - 17 November 2023 through 20 November 2023

ER -

Xue B, Zheng Q, Li Z, Zhao W, Zhang W, Feng X. Deep Federated Feature Recommendation. In Tallon-Ballesteros AJ, Cortes-Ancos E, Lopez-Garcia DA, editors, Electronics, Communications and Networks - Proceedings of the 13th International Conference, CECNet 2023. IOS Press BV. 2024. p. 644-649. (Frontiers in Artificial Intelligence and Applications). doi: 10.3233/FAIA231248

Deep Federated Feature Recommendation

Abstract

Publication series

Conference

Keywords

Access to Document

Other files and links

Fingerprint

Cite this