Variable Selection based on a Two-stage Projection Pursuit Algorithm

Shu Jiang; Yijun Xie

doi:10.5220/0009098901880193

Variable Selection based on a Two-stage Projection Pursuit Algorithm

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

Abstract

Dimension reduction methods have gained popularity in modern era due to exponential growth in data collection. Extracting key information and learning from all available data is a crucial step. Principal component analysis (PCA) is a popular dimension reduction technique due to its simplicity and flexibility. We stress that PCA is solely based on maximizing the proportion of total variance of the explanatory variables and do not directly impact the outcome of interest. Variable selection under such unsupervised setting may thus be inefficient. In this note, we propose a novel two-stage projection pursuit based algorithm which simultaneously consider the loss in the outcome variable when doing variable selection. We believe that when one is keen in variable selection in relation to the outcome of interest, the proposed method may be more efficient compared to existing methods.

Original language	English
Title of host publication	BIOINFORMATICS 2020 - 11th International Conference on Bioinformatics Models, Methods and Algorithms, Proceedings; Part of 13th International Joint Conference on Biomedical Engineering Systems and Technologies, BIOSTEC 2020
Editors	Elisabetta De Maria, Ana Fred, Hugo Gamboa
Publisher	SciTePress
Pages	188-193
Number of pages	6
ISBN (Electronic)	9789897583988
DOIs	https://doi.org/10.5220/0009098901880193
State	Published - 2020
Event	11th International Conference on Bioinformatics Models, Methods and Algorithms, BIOINFORMATICS 2020 - Part of 13th International Joint Conference on Biomedical Engineering Systems and Technologies, BIOSTEC 2020 - Valletta, Malta Duration: Feb 24 2020 → Feb 26 2020

Publication series

Name	BIOINFORMATICS 2020 - 11th International Conference on Bioinformatics Models, Methods and Algorithms, Proceedings; Part of 13th International Joint Conference on Biomedical Engineering Systems and Technologies, BIOSTEC 2020

Conference

Conference	11th International Conference on Bioinformatics Models, Methods and Algorithms, BIOINFORMATICS 2020 - Part of 13th International Joint Conference on Biomedical Engineering Systems and Technologies, BIOSTEC 2020
Country/Territory	Malta
City	Valletta
Period	02/24/20 → 02/26/20

Keywords

Optimization
Two-stage Projection Pursuit
Variable Selection

Access to Document

10.5220/0009098901880193

Cite this

Jiang, S., & Xie, Y. (2020). Variable Selection based on a Two-stage Projection Pursuit Algorithm. In E. De Maria, A. Fred, & H. Gamboa (Eds.), BIOINFORMATICS 2020 - 11th International Conference on Bioinformatics Models, Methods and Algorithms, Proceedings; Part of 13th International Joint Conference on Biomedical Engineering Systems and Technologies, BIOSTEC 2020 (pp. 188-193). (BIOINFORMATICS 2020 - 11th International Conference on Bioinformatics Models, Methods and Algorithms, Proceedings; Part of 13th International Joint Conference on Biomedical Engineering Systems and Technologies, BIOSTEC 2020). SciTePress. https://doi.org/10.5220/0009098901880193

Jiang, Shu ; Xie, Yijun. / Variable Selection based on a Two-stage Projection Pursuit Algorithm. BIOINFORMATICS 2020 - 11th International Conference on Bioinformatics Models, Methods and Algorithms, Proceedings; Part of 13th International Joint Conference on Biomedical Engineering Systems and Technologies, BIOSTEC 2020. editor / Elisabetta De Maria ; Ana Fred ; Hugo Gamboa. SciTePress, 2020. pp. 188-193 (BIOINFORMATICS 2020 - 11th International Conference on Bioinformatics Models, Methods and Algorithms, Proceedings; Part of 13th International Joint Conference on Biomedical Engineering Systems and Technologies, BIOSTEC 2020).

@inproceedings{0e48d7e2b0014a8880917df31b75a704,

title = "Variable Selection based on a Two-stage Projection Pursuit Algorithm",

abstract = "Dimension reduction methods have gained popularity in modern era due to exponential growth in data collection. Extracting key information and learning from all available data is a crucial step. Principal component analysis (PCA) is a popular dimension reduction technique due to its simplicity and flexibility. We stress that PCA is solely based on maximizing the proportion of total variance of the explanatory variables and do not directly impact the outcome of interest. Variable selection under such unsupervised setting may thus be inefficient. In this note, we propose a novel two-stage projection pursuit based algorithm which simultaneously consider the loss in the outcome variable when doing variable selection. We believe that when one is keen in variable selection in relation to the outcome of interest, the proposed method may be more efficient compared to existing methods.",

keywords = "Optimization, Two-stage Projection Pursuit, Variable Selection",

author = "Shu Jiang and Yijun Xie",

note = "Publisher Copyright: Copyright {\textcopyright} 2020 by SCITEPRESS – Science and Technology Publications, Lda. All rights reserved; 11th International Conference on Bioinformatics Models, Methods and Algorithms, BIOINFORMATICS 2020 - Part of 13th International Joint Conference on Biomedical Engineering Systems and Technologies, BIOSTEC 2020 ; Conference date: 24-02-2020 Through 26-02-2020",

year = "2020",

doi = "10.5220/0009098901880193",

language = "English",

series = "BIOINFORMATICS 2020 - 11th International Conference on Bioinformatics Models, Methods and Algorithms, Proceedings; Part of 13th International Joint Conference on Biomedical Engineering Systems and Technologies, BIOSTEC 2020",

publisher = "SciTePress",

pages = "188--193",

editor = "{De Maria}, Elisabetta and Ana Fred and Hugo Gamboa",

booktitle = "BIOINFORMATICS 2020 - 11th International Conference on Bioinformatics Models, Methods and Algorithms, Proceedings; Part of 13th International Joint Conference on Biomedical Engineering Systems and Technologies, BIOSTEC 2020",

}

Jiang, S & Xie, Y 2020, Variable Selection based on a Two-stage Projection Pursuit Algorithm. in E De Maria, A Fred & H Gamboa (eds), BIOINFORMATICS 2020 - 11th International Conference on Bioinformatics Models, Methods and Algorithms, Proceedings; Part of 13th International Joint Conference on Biomedical Engineering Systems and Technologies, BIOSTEC 2020. BIOINFORMATICS 2020 - 11th International Conference on Bioinformatics Models, Methods and Algorithms, Proceedings; Part of 13th International Joint Conference on Biomedical Engineering Systems and Technologies, BIOSTEC 2020, SciTePress, pp. 188-193, 11th International Conference on Bioinformatics Models, Methods and Algorithms, BIOINFORMATICS 2020 - Part of 13th International Joint Conference on Biomedical Engineering Systems and Technologies, BIOSTEC 2020, Valletta, Malta, 02/24/20. https://doi.org/10.5220/0009098901880193

Variable Selection based on a Two-stage Projection Pursuit Algorithm. / Jiang, Shu; Xie, Yijun.
BIOINFORMATICS 2020 - 11th International Conference on Bioinformatics Models, Methods and Algorithms, Proceedings; Part of 13th International Joint Conference on Biomedical Engineering Systems and Technologies, BIOSTEC 2020. ed. / Elisabetta De Maria; Ana Fred; Hugo Gamboa. SciTePress, 2020. p. 188-193 (BIOINFORMATICS 2020 - 11th International Conference on Bioinformatics Models, Methods and Algorithms, Proceedings; Part of 13th International Joint Conference on Biomedical Engineering Systems and Technologies, BIOSTEC 2020).

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

TY - GEN

T1 - Variable Selection based on a Two-stage Projection Pursuit Algorithm

AU - Jiang, Shu

AU - Xie, Yijun

PY - 2020

Y1 - 2020

N2 - Dimension reduction methods have gained popularity in modern era due to exponential growth in data collection. Extracting key information and learning from all available data is a crucial step. Principal component analysis (PCA) is a popular dimension reduction technique due to its simplicity and flexibility. We stress that PCA is solely based on maximizing the proportion of total variance of the explanatory variables and do not directly impact the outcome of interest. Variable selection under such unsupervised setting may thus be inefficient. In this note, we propose a novel two-stage projection pursuit based algorithm which simultaneously consider the loss in the outcome variable when doing variable selection. We believe that when one is keen in variable selection in relation to the outcome of interest, the proposed method may be more efficient compared to existing methods.

AB - Dimension reduction methods have gained popularity in modern era due to exponential growth in data collection. Extracting key information and learning from all available data is a crucial step. Principal component analysis (PCA) is a popular dimension reduction technique due to its simplicity and flexibility. We stress that PCA is solely based on maximizing the proportion of total variance of the explanatory variables and do not directly impact the outcome of interest. Variable selection under such unsupervised setting may thus be inefficient. In this note, we propose a novel two-stage projection pursuit based algorithm which simultaneously consider the loss in the outcome variable when doing variable selection. We believe that when one is keen in variable selection in relation to the outcome of interest, the proposed method may be more efficient compared to existing methods.

KW - Optimization

KW - Two-stage Projection Pursuit

KW - Variable Selection

UR - http://www.scopus.com/inward/record.url?scp=85083558650&partnerID=8YFLogxK

U2 - 10.5220/0009098901880193

DO - 10.5220/0009098901880193

M3 - Conference contribution

AN - SCOPUS:85083558650

T3 - BIOINFORMATICS 2020 - 11th International Conference on Bioinformatics Models, Methods and Algorithms, Proceedings; Part of 13th International Joint Conference on Biomedical Engineering Systems and Technologies, BIOSTEC 2020

SP - 188

EP - 193

BT - BIOINFORMATICS 2020 - 11th International Conference on Bioinformatics Models, Methods and Algorithms, Proceedings; Part of 13th International Joint Conference on Biomedical Engineering Systems and Technologies, BIOSTEC 2020

A2 - De Maria, Elisabetta

A2 - Fred, Ana

A2 - Gamboa, Hugo

PB - SciTePress

T2 - 11th International Conference on Bioinformatics Models, Methods and Algorithms, BIOINFORMATICS 2020 - Part of 13th International Joint Conference on Biomedical Engineering Systems and Technologies, BIOSTEC 2020

Y2 - 24 February 2020 through 26 February 2020

ER -

Jiang S, Xie Y. Variable Selection based on a Two-stage Projection Pursuit Algorithm. In De Maria E, Fred A, Gamboa H, editors, BIOINFORMATICS 2020 - 11th International Conference on Bioinformatics Models, Methods and Algorithms, Proceedings; Part of 13th International Joint Conference on Biomedical Engineering Systems and Technologies, BIOSTEC 2020. SciTePress. 2020. p. 188-193. (BIOINFORMATICS 2020 - 11th International Conference on Bioinformatics Models, Methods and Algorithms, Proceedings; Part of 13th International Joint Conference on Biomedical Engineering Systems and Technologies, BIOSTEC 2020). doi: 10.5220/0009098901880193

Variable Selection based on a Two-stage Projection Pursuit Algorithm

Abstract

Publication series

Conference

Keywords

Access to Document

Other files and links

Fingerprint

Cite this