Simple power series for pattern classification

Kar Ann Toh

doi:10.1109/ICIEA.2007.4318486

Simple power series for pattern classification

Kar Ann Toh

Department of Electrical and Electronic Engineering

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution

Abstract

We show in this paper, that a simple power series model with appropriate learning formulation, can be used for effective pattern classification. Essentially, an error counting cost function is adopted. Through a linear parametric power series model and a quadratic approximation to the error cost, a deterministic solution is derived. This solution is seen to relate to a class-specific setting of the more generic weighted least-squares. A tuning mechanism is thus incorporated for robust applications. Our empirical evaluations show effectiveness of the classifier.

Original language	English
Title of host publication	ICIEA 2007
Subtitle of host publication	2007 Second IEEE Conference on Industrial Electronics and Applications
Pages	641-646
Number of pages	6
DOIs	https://doi.org/10.1109/ICIEA.2007.4318486
Publication status	Published - 2007
Event	2007 2nd IEEE Conference on Industrial Electronics and Applications, ICIEA 2007 - Harbin, China Duration: 2007 May 23 → 2007 May 25

Publication series

Name	ICIEA 2007: 2007 Second IEEE Conference on Industrial Electronics and Applications

Other

Other	2007 2nd IEEE Conference on Industrial Electronics and Applications, ICIEA 2007
Country	China
City	Harbin
Period	07/5/23 → 07/5/25

All Science Journal Classification (ASJC) codes

Control and Systems Engineering
Electrical and Electronic Engineering

Access to Document

10.1109/ICIEA.2007.4318486

Fingerprint Dive into the research topics of 'Simple power series for pattern classification'. Together they form a unique fingerprint.

Cite this

@inproceedings{de81f0a5248747b094c0e5d1356e0aea,

title = "Simple power series for pattern classification",

abstract = "We show in this paper, that a simple power series model with appropriate learning formulation, can be used for effective pattern classification. Essentially, an error counting cost function is adopted. Through a linear parametric power series model and a quadratic approximation to the error cost, a deterministic solution is derived. This solution is seen to relate to a class-specific setting of the more generic weighted least-squares. A tuning mechanism is thus incorporated for robust applications. Our empirical evaluations show effectiveness of the classifier.",

author = "Toh, {Kar Ann}",

note = "Copyright: Copyright 2008 Elsevier B.V., All rights reserved.; 2007 2nd IEEE Conference on Industrial Electronics and Applications, ICIEA 2007 ; Conference date: 23-05-2007 Through 25-05-2007",

year = "2007",

doi = "10.1109/ICIEA.2007.4318486",

language = "English",

isbn = "1424407370",

series = "ICIEA 2007: 2007 Second IEEE Conference on Industrial Electronics and Applications",

pages = "641--646",

booktitle = "ICIEA 2007",

}

Toh, KA 2007, Simple power series for pattern classification. in ICIEA 2007: 2007 Second IEEE Conference on Industrial Electronics and Applications., 4318486, ICIEA 2007: 2007 Second IEEE Conference on Industrial Electronics and Applications, pp. 641-646, 2007 2nd IEEE Conference on Industrial Electronics and Applications, ICIEA 2007, Harbin, China, 07/5/23. https://doi.org/10.1109/ICIEA.2007.4318486

TY - GEN

T1 - Simple power series for pattern classification

AU - Toh, Kar Ann

PY - 2007

Y1 - 2007

N2 - We show in this paper, that a simple power series model with appropriate learning formulation, can be used for effective pattern classification. Essentially, an error counting cost function is adopted. Through a linear parametric power series model and a quadratic approximation to the error cost, a deterministic solution is derived. This solution is seen to relate to a class-specific setting of the more generic weighted least-squares. A tuning mechanism is thus incorporated for robust applications. Our empirical evaluations show effectiveness of the classifier.

AB - We show in this paper, that a simple power series model with appropriate learning formulation, can be used for effective pattern classification. Essentially, an error counting cost function is adopted. Through a linear parametric power series model and a quadratic approximation to the error cost, a deterministic solution is derived. This solution is seen to relate to a class-specific setting of the more generic weighted least-squares. A tuning mechanism is thus incorporated for robust applications. Our empirical evaluations show effectiveness of the classifier.

UR - http://www.scopus.com/inward/record.url?scp=35248812568&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=35248812568&partnerID=8YFLogxK

U2 - 10.1109/ICIEA.2007.4318486

DO - 10.1109/ICIEA.2007.4318486

M3 - Conference contribution

AN - SCOPUS:35248812568

SN - 1424407370

SN - 9781424407378

T3 - ICIEA 2007: 2007 Second IEEE Conference on Industrial Electronics and Applications

SP - 641

EP - 646

BT - ICIEA 2007

T2 - 2007 2nd IEEE Conference on Industrial Electronics and Applications, ICIEA 2007

Y2 - 23 May 2007 through 25 May 2007

ER -