Electrostatic field classifier for deficient data

This source preferred by Marcin Budka

Authors: Budka, M. and Gabrys, B.

Editors: Kurzynski, M. and Wozniak, M.

http://eprints.bournemouth.ac.uk/9541/

http://springerlink.com/content/5148221168341945/?p=01aabf3130d945fe83aaf774c00360bb&pi=8

Pages: 311-318

Publisher: Springer

Place of Publication: Heidelberg

ISBN: 978-3-540-93904-7

DOI: 10.1007/978-3-540-93905-4_37

This paper investigates the suitability of recently developed models based on the physical field phenomena for classification problems with incomplete datasets. An original approach to exploiting incomplete training data with missing features and labels, involving extensive use of electrostatic charge analogy, has been proposed. Classification of incomplete patterns has been investigated using a local dimensionality reduction technique, which aims at exploiting all available information rather than trying to estimate the missing values. The performance of all proposed methods has been tested on a number of benchmark datasets for a wide range of missing data scenarios and compared to the performance of some standard techniques. Several modifications of the original electrostatic field classifier aiming at improving speed and robustness in higher dimensional spaces are also discussed.

This data was imported from Scopus:

Authors: Budka, M. and Gabrys, B.

http://eprints.bournemouth.ac.uk/9541/

Journal: Advances in Intelligent and Soft Computing

Volume: 57

Pages: 311-318

eISSN: 1860-0794

ISSN: 1867-5662

© Springer-Verlag Berlin Heidelberg 2009. This paper investigates the suitability of recently developed models based on the physical field phenomena for classification of incomplete datasets. An original approach to exploiting incomplete training data with missing features and labels, involving extensive use of electrostatic charge analogy has been proposed. Classification of incomplete patterns has been investigated using a local dimensionality reduction technique, which aims at exploiting all available information rather than trying to estimate the missing values. The performance of all proposed methods has been tested on a number of benchmark datasets for a wide range of missing data scenarios and compared to the performance of some standard techniques.

The data on this page was last updated at 17:31 on November 21, 2017.