A General Framework for Dimensionality-Reducing Data Visualization Mapping
- 1 March 2012
- journal article
- Published by MIT Press in Neural Computation
- Vol. 24 (3) , 771-804
Abstract
In recent years, a wealth of dimension-reduction techniques for data visualization and preprocessing has been established. Nonparametric methods require additional effort for out-of-sample extensions, because they provide only a mapping of a given finite set of points. In this letter, we propose a general view on nonparametric dimension reduction based on the concept of cost functions and properties of the data. Based on this general principle, we transfer nonparametric dimension reduction to explicit mappings of the data manifold such that direct out-of-sample extensions become possible. Furthermore, this concept offers the possibility of investigating the generalization ability of data visualization to new data points. We demonstrate the approach based on a simple global linear mapping, as well as prototype-based local linear mappings. In addition, we can bias the functional form according to given auxiliary information. This leads to explicit supervised visualization mappings with discriminative properties comparable to state-of-the-art approaches.Keywords
This publication has 25 references indexed in Scilit:
- Adaptive local dissimilarity measures for discriminative dimension reduction of labeled dataNeurocomputing, 2010
- Adaptive Relevance Matrices in Learning Vector QuantizationNeural Computation, 2009
- Quality assessment of dimensionality reduction: Rank-based criteriaNeurocomputing, 2009
- Parametric Embedding for Class VisualizationNeural Computation, 2007
- Kernel clustering-based discriminant analysisPattern Recognition, 2007
- Batch and median neural gasNeural Networks, 2006
- An extension of neural gas to local PCANeurocomputing, 2004
- Bankruptcy analysis with self-organizing maps in learning metricsIEEE Transactions on Neural Networks, 2001
- A Global Geometric Framework for Nonlinear Dimensionality ReductionScience, 2000
- 10.1162/153244303321897690Applied Physics Letters, 2000