FedSCOPE: A Comprehensive Evaluation Framework for Federated Learning in Human-Centered Social Computing
文献类型：期刊论文
| 作者 | Li, Yanli1,2; Li, Yuqi5; Zhou, Yanan4; Zhang, Yuning4; Yang, Nan4; Yuan, Dong4; Ding, Weiping2,3 |
| 刊名 | IEEE TRANSACTIONS ON COMPUTATIONAL SOCIAL SYSTEMS
 |
| 出版日期 | 2025-10-17 |
| 页码 | 16 |
| 关键词 | Evaluation metric federated learning (FL) model selection social computing |
| ISSN号 | 2329-924X |
| DOI | 10.1109/TCSS.2025.3603719 |
| 英文摘要 | Federated learning (FL) offers innovative solutions for training complex neural networks on large-scale data without directly accessing participants' raw data. Owing to its inherent privacy-preserving nature, FL has seen widespread deployment in real-world applications, particularly in social computing. While promising, existing performance evaluations are often one-dimensional and industrial-centric, failing to reflect the human-centered requirements of social computing scenarios. Consequently, how to comprehensively evaluate an FL algorithm and identify the most suitable candidate for social computing remains an open challenge. To mitigate the research gap, we introduce the federated learning social computing performance evaluation (FedSCOPE) framework in this study. FedSCOPE comprises five compulsory components-learning performance, reliability, fairness, robustness, and privacy preservation-that reflect the primary needs of participants in social computing FL systems, and includes an optional personalization component when required. Each component is weighted by an importance factor, and their integration yields a single FedSCOPE index that provides a holistic assessment of an FL algorithm. Three representative case studies in social computing were conducted to evaluate the effectiveness of the proposed FedSCOPE. PriHFLRw, FedAvgRw, and FLTrust achieved FedSCOPE scores of 98.7, 87.5, and 87.73, respectively, and were recommended as suitable algorithms for delivering smart services in the simulated scenarios. We hope this work can offer practical insights and guide the evaluation of FL algorithms in real-world applications. |
| 资助项目 | National Key RD Plan of China[2024YFE0202700] ; National Natural Science Foundation of China[U2433216] ; Natural Science Foundation of Jiangsu Province[BK20231337] ; Natural Science Foundation of Jiangsu Higher Education Institutions of China[24KJB520032] |
| WOS研究方向 | Computer Science |
| 语种 | 英语 |
| WOS记录号 | WOS:001596957500001 |
| 出版者 | IEEE-INST ELECTRICAL ELECTRONICS ENGINEERS INC |
| 源URL | [http://119.78.100.204/handle/2XEOYT63/41636]  |
| 专题 | 中国科学院计算技术研究所期刊论文_英文 |
| 通讯作者 | Ding, Weiping |
| 作者单位 | 1.Univ Sydney, Sch Elect & Comp Engn, Sydney, NSW 2006, Australia 2.Nantong Univ, Sch Artificial Intelligence & Comp Sci, Nantong 226019, Peoples R China 3.City Univ Macau, Fac Data Sci, Taipa 999078, Macau, Peoples R China 4.Univ Sydney, Sch Elect & Comp Engn, Sydney, NSW 2006, Australia 5.Chinese Acad Sci, Inst Comp Technol, Beijing 100864, Peoples R China |
| 推荐引用方式 GB/T 7714 | Li, Yanli,Li, Yuqi,Zhou, Yanan,et al. FedSCOPE: A Comprehensive Evaluation Framework for Federated Learning in Human-Centered Social Computing[J]. IEEE TRANSACTIONS ON COMPUTATIONAL SOCIAL SYSTEMS,2025:16. |
| APA | Li, Yanli.,Li, Yuqi.,Zhou, Yanan.,Zhang, Yuning.,Yang, Nan.,...&Ding, Weiping.(2025).FedSCOPE: A Comprehensive Evaluation Framework for Federated Learning in Human-Centered Social Computing.IEEE TRANSACTIONS ON COMPUTATIONAL SOCIAL SYSTEMS,16. |
| MLA | Li, Yanli,et al."FedSCOPE: A Comprehensive Evaluation Framework for Federated Learning in Human-Centered Social Computing".IEEE TRANSACTIONS ON COMPUTATIONAL SOCIAL SYSTEMS (2025):16. |
入库方式： OAI收割
来源：计算技术研究所
浏览0
下载0
收藏0
其他版本
除非特别说明，本系统中所有内容都受版权保护，并保留所有权利。