Please use this identifier to cite or link to this item:
https://hdl.handle.net/2440/72133
Citations | ||
Scopus | Web of Science® | Altmetric |
---|---|---|
?
|
?
|
Type: | Conference paper |
Title: | On the optimality of sequential forward feature selection using class separability measure |
Author: | Wang, L. Shen, C. Hartley, R. |
Citation: | Proceedings of the International Conference on Digital Image Computing Techniques and Applications (DICTA'11), held in Noosa, Queensland, 6-8 December, 2011: pp.203-208 |
Publisher: | IEEE |
Publisher Place: | USA |
Issue Date: | 2011 |
ISBN: | 9781457720062 |
Conference Name: | International Conference on Digital Image Computing Techniques and Applications (2011 : Noosa, Qld) |
Statement of Responsibility: | Lei Wang, Chunhua Shen and Richard Hartley |
Abstract: | This paper studies sequential forward feature selection that uses the scatter-matrix-based class separability measure. We find that by adding a scale factor to each iteration of the conventional sequential selection, a sequential selection that guarantees the global optimum can be attained. We give a thorough theoretical proof of its optimality via a novel geometric interpretation, and this leads to a unified framework including the optimal sequential selection, the conventional sequential selection and the best-individual-N selection. In addition, we show that with our formulation, feature selection can be treated as a linear fractional maximization problem, and it can be efficiently solved by algorithms well developed in the literature. This gives a non-sequential globally optimal feature selection algorithm. Both theoretical and experimental study demonstrate their efficiency. |
Keywords: | Sequential feature selection class separability |
Rights: | © 2011 IEEE |
DOI: | 10.1109/DICTA.2011.41 |
Published version: | http://dx.doi.org/10.1109/dicta.2011.41 |
Appears in Collections: | Aurora harvest Computer Science publications |
Files in This Item:
There are no files associated with this item.
Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.