Extension of Incremental Linear Discriminant Analysis to Online Feature Extraction under Nonstationary Environments

Extension of Incremental Linear Discriminant Analysis to Online Feature Extraction under Nonstationary Environments

In this paper, a new approach to an online feature extrac�tion under nonstationary environments is proposed by extending Incre�mental Linear Discriminant Analysis (ILDA). The extended ILDA not only detect so-called “concept drifts” but also transfer the knowledge on discriminant feature spaces of th...

Full description

Saved in:
Bibliographic Details
Main Authors: Annie, Joseph, Young Min, Jang, Seiichi, Ozawa, Minho, Lee
Other Authors: Tingwen, Huang
Format: Book Chapter
Language:English
Published: Springer 2012
Subjects:
TK Electrical engineering. Electronics Nuclear engineering
TN Mining engineering. Metallurgy
Online Access:http://ir.unimas.my/id/eprint/39677/3/Neural%20Information%20Processing%20-%20Copy.pdf
http://ir.unimas.my/id/eprint/39677/
Tags: Add Tag
No Tags, Be the first to tag this record!
Institution: Universiti Malaysia Sarawak
Language: English
Description
Summary:In this paper, a new approach to an online feature extrac�tion under nonstationary environments is proposed by extending Incre�mental Linear Discriminant Analysis (ILDA). The extended ILDA not only detect so-called “concept drifts” but also transfer the knowledge on discriminant feature spaces of the past concepts to construct good feature spaces. The performance of the extended ILDA is evaluated for the benchmark datasets including sudden changes and reoccurrence in concepts.

Similar Items