Semi-supervised geodesic Generative Topographic Mapping

Authors:
Raúl Cruz-Barbosa;Alfredo Vellido
Affiliations:
Technical University of Catalonia, 08034 Barcelona, Spain and Technological University of the Mixteca, 69000 Huajuapan, Oaxaca, Mexico;Technical University of Catalonia, 08034 Barcelona, Spain
Venue:
Pattern Recognition Letters
Year:
2010

Citing 5
Cited 5

Algorithms for clustering data

Algorithms for clustering data
GTM: the generative topographic mapping

Neural Computation
Semi-Supervised Learning on Riemannian Manifolds

Machine Learning
A Semi-supervised SVM for Manifold Learning

ICPR '06 Proceedings of the 18th International Conference on Pattern Recognition - Volume 02
Manifold constrained finite gaussian mixtures

IWANN'05 Proceedings of the 8th international conference on Artificial Neural Networks: computational Intelligence and Bioinspired Systems

A variational Bayesian approach for the robust analysis of the cortical silent period from EMG recordings of brain stroke patients

Neurocomputing
Comparative diagnostic accuracy of linear and nonlinear feature extraction methods in a neuro-oncology problem

MCPR'11 Proceedings of the Third Mexican conference on Pattern recognition
On the computation of the geodesic distance with an application to dimensionality reduction in a neuro-oncology problem

CIARP'11 Proceedings of the 16th Iberoamerican Congress conference on Progress in Pattern Recognition, Image Analysis, Computer Vision, and Applications
Cartogram visualization for nonlinear manifold learning models

Data Mining and Knowledge Discovery
A multi-manifold semi-supervised Gaussian mixture model for pattern classification

Pattern Recognition Letters

Quantified Score

Hi-index	0.10

Visualization

Abstract

We present a novel semi-supervised model, SS-Geo-GTM, which stems from a geodesic distance-based extension of Generative Topographic Mapping that prioritizes neighbourhood relationships along a generated manifold embedded in the observed data space. With this, it improves the trustworthiness and the continuity of the low-dimensional representations it provides, while behaving robustly in the presence of noise. In SS-Geo-GTM, the model prototypes are linked by the nearest neighbour to the data manifold constructed by Geo-GTM. The resulting proximity graph is used as the basis for a class label propagation algorithm. The performance of SS-Geo-GTM is experimentally assessed, comparing positively with that of an Euclidean distance-based counterpart and with those of alternative manifold learning methods.