Generalized topographic block model
Generalized topographic block model
Co-clustering leads to parsimony in data visualisation with a number of parameters dramatically reduced in comparison to the dimensions of the data sample. Herein, we propose a new generalized approach for nonlinear mapping by a re-parameterization of the latent block mixture model. The densities modeling the blocks are in an exponential family such that the Gaussian, Bernoulli and Poisson laws are particular cases. The inference of the parameters is derived from the block expectation–maximization algorithm with a Newton–Raphson procedure at the maximization step. Empirical experiments with textual data validate the interest of our generalized model.
latent block mixture model, exponential family, generative topographic mapping, block expectation–maximization, visualisation
1-8
Priam, Rodolphe
f9a5a1b9-fe9e-4f82-bef2-6f0b97de1673
Nadif, Mohamed
7d711a92-f0f6-49b2-8d12-b09c8ed41124
Govaert, Gérard
d7585fe7-bba1-4f53-9d95-2c6382fa7183
Priam, Rodolphe
f9a5a1b9-fe9e-4f82-bef2-6f0b97de1673
Nadif, Mohamed
7d711a92-f0f6-49b2-8d12-b09c8ed41124
Govaert, Gérard
d7585fe7-bba1-4f53-9d95-2c6382fa7183
Abstract
Co-clustering leads to parsimony in data visualisation with a number of parameters dramatically reduced in comparison to the dimensions of the data sample. Herein, we propose a new generalized approach for nonlinear mapping by a re-parameterization of the latent block mixture model. The densities modeling the blocks are in an exponential family such that the Gaussian, Bernoulli and Poisson laws are particular cases. The inference of the parameters is derived from the block expectation–maximization algorithm with a Newton–Raphson procedure at the maximization step. Empirical experiments with textual data validate the interest of our generalized model.
Text
__userfiles.soton.ac.uk_Library_SLAs_Work_for_ALL's_Work_for_ePrints_Accepted Manuscripts_Priam_Generalized.pdf
- Accepted Manuscript
More information
Accepted/In Press date: 8 April 2015
e-pub ahead of print date: 8 September 2015
Keywords:
latent block mixture model, exponential family, generative topographic mapping, block expectation–maximization, visualisation
Organisations:
Statistical Sciences Research Institute
Identifiers
Local EPrints ID: 383632
URI: http://eprints.soton.ac.uk/id/eprint/383632
ISSN: 0925-2312
PURE UUID: 9cfd4737-551c-418d-bedf-d420facb1853
Catalogue record
Date deposited: 09 Nov 2015 16:34
Last modified: 14 Mar 2024 21:45
Export record
Altmetrics
Contributors
Author:
Rodolphe Priam
Author:
Mohamed Nadif
Author:
Gérard Govaert
Download statistics
Downloads from ePrints over the past year. Other digital versions may also be available to download e.g. from the publisher's website.
View more statistics