The views expressed here are those of the individual authors and not necessarily those of the JSM sponsors, their officers, or their staff.
Online Program Home
Abstract Details
|
Activity Number:
|
253
|
|
Type:
|
Contributed
|
|
Date/Time:
|
Monday, July 30, 2012 : 2:00 PM to 3:50 PM
|
|
Sponsor:
|
Biometrics Section
|
| Abstract - #304661 |
|
Title:
|
Modified Linear Discriminant Analysis
|
|
Author(s):
|
Jin Hyun Nam*+ and Donguk Kim
|
|
Companies:
|
and Sungkyunkwan University
|
|
Address:
|
123-301 Hannam-The Hill APT., Seoul, _, 82, South Korea
|
|
Keywords:
|
Modified LDA ;
DLDA ;
Classification ;
Gene expression
|
|
Abstract:
|
Many classification methods has been applied for microarray data to compare the performance. Among these, linear discriminant analysis (LDA) is a favored tool due to its simplicity, robustness and predictive accuracy but when the number of genes is larger than the number of observations, it can not be applied directly because the within-class covariance matrix of the genes is singular. Also, diagonal LDA (DLDA) is simpler model compared to LDA and has better performance in some cases. However, in reality, DLDA requires a strong assumption based on conditional independence. In this paper, we propose the modified LDA (MLDA). MLDA is based on conditional independence but uses the information that has an effect on classification performance with joint probability. We suggest two approaches. One is the case of using gene selection method. The other is no use of gene selection method. We found that MLDA has better performance than LDA or DLDA and is no worse than KNN or SVM in real data analysis.
|
The address information is for the authors that have a + after their name.
Authors who are presenting talks have a * after their name.
Back to the full JSM 2012 program
|
2012 JSM Online Program Home
For information, contact jsm@amstat.org or phone (888) 231-3473.
If you have questions about the Continuing Education program, please contact the Education Department.