Visualizing the quality of dimensionality reduction

NEUROCOMPUTING

卷:112

Article

Mokbel, Bassam¹  Lueks, Wouter^2,3  Gisbrecht, Andrej¹  Hammer, Barbara¹ 

[1] Univ Bielefeld, CITEC Ctr Excellence, Bielefeld, Germany

[2] Univ Groningen, Fac Math & Nat Sci, NL-9700 AB Groningen, Netherlands

[3] Univ Nijmegen, Fac Sci, Nijmegen, Netherlands

关键词: Nonlinear dimensionality reduction;    Data visualization;    Quality assessment;    Co-ranking matrix;   

DOI  :  10.1016/j.neucom.2012.11.046

来源: Elsevier

【 摘 要 】

The growing number of dimensionality reduction methods available for data visualization has recently inspired the development of formal measures to evaluate the resulting low-dimensional representation independently from the methods' inherent criteria. Many evaluation measures can be summarized based on the co-ranking matrix. In this work, we analyze the characteristics of the co-ranking framework, focusing on interpretability and controllability in evaluation scenarios where a fine-grained assessment of a given visualization is desired. We extend the framework in two ways: (i) we propose how to link the evaluation to point-wise quality measures which can be used directly to augment the evaluated visualization and highlight erroneous regions; (ii) we improve the parameterization of the quality measure to offer more direct control over the evaluation's focus, and thus help the user to investigate more specific characteristics of the visualization. (C) 2013 Elsevier B.V. All rights reserved.

【 授权许可】

Free