Search papers, labs, and topics across Lattice.
This paper introduces a novel clustering algorithm tailored for data residing on the unit hypersphere, leveraging the d-dimensional generalized Kuramoto model to capture the sphere's geometric structure. The algorithm synchronizes data points on the hypersphere, grouping them based on their phase relationships within the Kuramoto model. Empirical evaluations on synthetic and real-world datasets demonstrate that the proposed method achieves comparable or superior clustering accuracy compared to traditional methods.
A synchronization-based approach provides a new way to cluster data on hyperspheres, outperforming standard methods by better respecting the data's inherent geometry.
Clustering on the unit hypersphere is a fundamental problem in various fields, with applications ranging from gene expression analysis to text and image classification. Traditional clustering methods are not always suitable for unit sphere data, as they do not account for the geometric structure of the sphere. We introduce a novel algorithm for clustering data represented as points on the unit sphere $\mathbf{S}^{d-1}$. Our method is based on the $d$-dimensional generalized Kuramoto model. The effectiveness of the introduced method is demonstrated on synthetic and real-world datasets. Results are compared with some of the traditional clustering methods, showing that our method achieves similar or better results in terms of clustering accuracy.
