In plain English
A federated approach to density-based clustering (a method that groups data points by how densely they are packed together) removes the need to manually tune key parameters while keeping raw data local on each device. Applied to image recognition, the method shows how privacy-preserving distributed learning can work without the trial-and-error setup that typically burdens such algorithms.
Year: 2026
Venue: PLoS ONE
Type: journal
DOI: 10.1371/journal.pone.0355161
External link: https://doi.org/10.1371/journal.pone.0355161
Abstract
DBSCAN (A Density-Based Algorithm for Discovering Clusters in Spatial Databases with Noise) is a classic clustering algorithm. However, clustering distributed data with privacy protection in edge computing environments is a key challenge for DBSCAN. In this research, we combine federated clustering and DBSCAN and propose two secure federated parameter-free DBSCAN clustering methods, called FDBSCAN and FDBSCAN++. The process involves the following steps: (1) differential privacy is applied to the client data and adaptive DBSCAN is used at each client to identify core points; (2) the clients send the extracted core points to the server, where the server aggregates these to obtain the final global cluster centers (FDBSCAN and FDBSCAN++ use different methods in this step); (3) the final clusters are generated using these global centers. To verify the effectiveness of the proposed two algorithms, we use eight real datasets, including the large-scale image dataset MNIST. Compared with traditional and state-of-the-art (SOTA) improved DBSCAN and federated clustering algorithms, the proposed algorithms achieve better clustering accuracy. In addition, we also apply FDBSCAN++ to image clustering and segmentation tasks, which achieves satisfactory results.