Distributed sketched subspace clustering for large-scale datasets
Panagiotis A. Traganitis, Georgios B. Giannakis · 2017
Subspace clustering has been a successful tool for unsupervised classification of high-dimensional and generally non linearly separable data. However, state-of-the-art subspace clustering algorithms do not scale well as the number of data increases. The present paper puts forth a distributed subspace clustering scheme for high-volume data based on random projections. Additionally, the method can cope with corrupted data. Performance of the novel scheme is assessed via numerical tests, and is compared with state-of-the-art subspace clustering methods.