"Dataset Title TPP-FL-CPS: A Dataset for Trustworthy Privacy-Preserving Federated Learning in Cyber-Physical Systems"
Abstract
"This dataset supports research on trustworthy privacy-preserving federated learning for cyber-physical systems. It provides data for federated clients, CPS sensor observations, algorithmic performance comparison, privacy leakage analysis, poisoning robustness, communication scalability, trust-score evolution, non-IID client partitioning, ablation analysis, and figure-source mapping. The dataset was designed to reproduce the experimental tables and figures used in the manuscript entitled \u201cTrustworthy Privacy-Preserving Federated Learning for Secure and Intelligent Cyber-Physical Systems Across Critical Infrastructure Environments.\u201d The dataset contains no personal, patient-level, proprietary industrial, or real critical-infrastructure data.The dataset contains ten CSV files and one Excel summary file. It includes 100 simulated federated clients across transportation, industrial IoT, smart energy, and healthcare CPS domains. It also includes 5,000 CPS sensor records, 42 algorithm-comparison records, 42 privacy-attack records, 294 poisoning-robustness records, 56 communication-scalability records, 6,000 trust-score evolution records, 100 non-IID partition records, 6 ablation-study records, and 16 figure-source mapping entries."
Read the paper
Related studies
- Active Learning during Federated Learning for Object Detection
- Fairness-aware federated learning
- Decentralized Federated Learning
- Building a Personally Identifiable Information Recognizer in a Privacy Preserved Manner Using Automated Annotation and Federated Learning
- Preservation of Patient Level Privacy: Federated Classification and Calibration Models
- Efficient explainable federated learning
Go deeper in Evidence Hub
Evidence Hub adds a study data sheet (design, sample size, evidence level, key numbers), an AI summary and a chat grounded in the paper's text.