A novel filter algorithm for unsupervised feature selection based on a space filling measure

Details

Ressource 1Download: es2018-57.pdf (2302.68 [Ko])
State: Public
Version: Final published version
Serval ID
serval:BIB_3B1EF6BC8A0C
Type
Proceedings: the proceedings of a conference.
Collection
Publications
Institution
Title
A novel filter algorithm for unsupervised feature selection based on a space filling measure
Organization
26rd European Symposium on Artificial Neural Networks, Computational Intelligence and Machine Learning (ESANN)
Address
Bruges (Belgium)
ISBN
978-287587047-6
Issued date
2018
Editor
Laib Mohamed, Kanevski Mikhail
Number of pages
485
Abstract
The research proposes a novel filter algorithm for the unsupervised feature selection problems based on a space filling measure. A well-known criterion of space filling design, called the coverage measure, is adapted to dimensionality reduction problems. Originally, this measure was developed to judge the quality of a space filling design. In this work it is used to reduce the redundancy in data. The proposed algorithm is evaluated on simulated data with several scenarios of noise injection. Furthermore, a comparison with some benchmark methods of feature selection is performed on real UCI datasets.
Keywords
Feature selection, Space-filling design, data mining, machine learning
Create date
17/07/2018 6:33
Last modification date
21/08/2019 6:08
Usage data