Spectral clustering using PCKID ? A probabilistic cluster kernel for incomplete data

Publication details

Journal: Lecture Notes in Computer Science (LNCS), vol. 10269 LNCS, p. 431–442, 2017
Publisher: Springer
International Standard Numbers:
- Printed: 0302-9743
- Electronic: 1611-3349
Links:
- ARKIV: http://hdl.handle.net/10037/13697
- DOI: doi.org/10.1007/978-3-319-59126-1_36

In this paper, we propose PCKID, a novel, robust, kernel function for spectral clustering, specifically designed to handle incomplete data. By combining posterior distributions of Gaussian Mixture Models for incomplete data on different scales, we are able to learn a kernel for incomplete data that does not depend on any critical hyperparameters, unlike the commonly used RBF kernel. To evaluate our method, we perform experiments on two real datasets. PCKID outperforms the baseline methods for all fractions of missing values and in some cases outperforms the baseline methods with up to 25% points.