Normalized Cuts and Image Segmentation
IEEE Transactions on Pattern Analysis and Machine Intelligence
On clusterings-good, bad and spectral
FOCS '00 Proceedings of the 41st Annual Symposium on Foundations of Computer Science
Document Clustering Using Locality Preserving Indexing
IEEE Transactions on Knowledge and Data Engineering
Label propagation through linear neighborhoods
ICML '06 Proceedings of the 23rd international conference on Machine learning
A tutorial on spectral clustering
Statistics and Computing
| Hi-index | 0.00 |
In this paper, we propose a novel spectral clustering algorithm called: Locality Spectral Clustering (Lsc ) which assumes that each data point can be linearly reconstructed from its local neighborhoods. The Lsc algorithm firstly try to learn a smooth enough manifold structure on the data manifold and then computes the eigenvectors on the smooth manifold structure, then as former spectral clustering methods, we use the eigenvectors to help the k -means algorithm to do clustering. Experiments have been performed on toy data sets and real world data sets and have shown that our algorithm can effectively discover the cluster structure and holds much better clustering accuracy than former methods. It is also worth noting that our algorithm is also much more stable in parameter than former spectral clustering methods.