DC-PFL: A dynamic clustering-based personalized federated learning method for human activity recognition
Bibliographic record
Abstract
Human Activity Recognition (HAR) is essential in pervasive computing, healthcare, and human–computer interaction, where accurate interpretation of motion data underpins intelligent decision-making. Federated Learning (FL) enables privacy-preserving model training across distributed clients without sharing raw data, but suffers from degraded performance under Non-Independent and Identically Distributed (Non-IID) data, a common challenge in HAR due to user diversity and device heterogeneity. To address this, Personalized Federated Learning (PFL) introduces client-specific modeling, often via clustering. However, most existing approaches adopt static clustering strategies, lacking adaptability to dynamic changes in client data distributions. In this work, we propose DC-PFL, a Dynamic Clustering-based Personalized Federated Learning framework that performs round-wise client clustering using lightweight statistical features, like Average Peak Frequency (APF), percentiles, and Median Absolute Deviation (MAD) derived from local model parameters. This design ensures efficient and privacy-preserving similarity estimation across clients. By dynamically adjusting clusters during training, DC-PFL enables fine-grained personalization, better generalization, and improved robustness to Non-IID conditions. Experimental results on HAR benchmarks demonstrate that DC-PFL achieves superior performance in both accuracy and convergence speed compared to existing methods, including FedCHAR and standard FL baselines, validating its effectiveness in real-world federated HAR scenarios.
Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.
How this classification was reachedexpand
Full frame machine prediction
Teacher imitationNot calibrated prevalence, not ground truth. Human validation pending. The Gemma side is a direct model label for every work in the frame, read from the title-only record. The Codex side is a classifier learned from the 10,348 direct Codex labels and calibrated to design-weighted sample rates; fields without enough sample support carry no Codex call. Candidate is the union of the two sides; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels.
Distilled classifier scores by category (both heads)
| Category | Codex | Gemma |
|---|---|---|
| Metaresearch | 0.001 | 0.004 |
| Meta-epidemiology (narrow) | 0.001 | 0.000 |
| Meta-epidemiology (broad) | 0.002 | 0.001 |
| Bibliometrics | 0.001 | 0.001 |
| Science and technology studies | 0.001 | 0.001 |
| Scholarly communication | 0.001 | 0.002 |
| Open science | 0.003 | 0.002 |
| Research integrity | 0.001 | 0.001 |
| Insufficient payload (model declined to judge) | 0.002 | 0.001 |
Machine scores (provisional)
The two teacher heads of the student model, read on this work. A score orders the frame for review; it never asserts a category, and the validation status ships verbatim with every row.
Baseline scores from an immature model (maturity gate not passed, 7 training rounds). Scores rank; they never assert a category.
score_only:v0-immature-baseline · verbatim from the scoring run: score_only means the number may rank works, and no category label ships from itClassification
machine, unvalidatedMachine predicted; a candidate call from one source (direct Gemma or distilled Codex), not a consensus.
How this classification was reached, model by model and score by score, is at the end of the page under "How this classification was reached".