Article ID | Journal | Published Year | Pages | File Type |
---|---|---|---|---|
5129419 | Journal of Multivariate Analysis | 2017 | 18 Pages |
Abstract
We consider the problem of clustering functional data while jointly selecting the most relevant features for classification. Functional sparse clustering is here analytically defined as a variational problem with a hard thresholding constraint ensuring the sparsity of the solution. First, a unique solution to sparse clustering with hard thresholding in finite dimensions is proved to exist. Then, the infinite-dimensional generalization is given and proved to have a unique solution under reasonable assumptions. Both the multivariate and the functional versions of sparse clustering with hard thresholding exhibit improvements on other standard and sparse clustering strategies on simulated data. A real functional data application is also shown.
Related Topics
Physical Sciences and Engineering
Mathematics
Numerical Analysis
Authors
Davide Floriello, Valeria Vitelli,