A complete and improved mathematical framework of the geometric approach based on the scaled convex hull (SCH) and its application to large medical data sets

Hao Yan Guo; Yuan Zhi Cheng; Da Zheng Wang; Jia Cheng Xu

doi:10.4028/www.scientific.net/AMM.556-562.3935

A complete and improved mathematical framework of the geometric approach based on the scaled convex hull (SCH) and its application to large medical data sets

Hao Yan Guo
, Yuan Zhi Cheng
, Da Zheng Wang
, Jia Cheng Xu

School of Computer Science and Technology (School of Software)

Harbin Institute of Technology
School of Mathematics, Harbin Institute of Technology

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

Abstract

In this paper a complete and improved mathematical framework of the geometric approach based on the Scaled convex hull (SCH), which includes two parts, SCH-based geometric algorithms for SVM and a fast and novel geometric method for model selection in SCH-based SVM, is developed to solve SCH-based SVM classification of large data sets. On the basis of this framework, these geometric algorithms are more suitable for classification task of large data sets. Results of numerical experiments show that the proposed geometric algorithms can reduce kernel calculations and display nice performances on large medical data sets, such as computer assisted screening for lung cancer.

Original language	English
Title of host publication	Mechatronics Engineering, Computing and Information Technology
Publisher	Trans Tech Publications
Pages	3935-3940
Number of pages	6
ISBN (Print)	9783038351153
DOIs	https://doi.org/10.4028/www.scientific.net/AMM.556-562.3935
State	Published - 2014
Event	2014 International Conference on Mechatronics Engineering and Computing Technology, ICMECT 2014 - Shanghai, China Duration: 9 Apr 2014 → 10 Apr 2014

Publication series

Name	Applied Mechanics and Materials
Volume	556-562
ISSN (Print)	1660-9336
ISSN (Electronic)	1662-7482

Conference

Conference	2014 International Conference on Mechatronics Engineering and Computing Technology, ICMECT 2014
Country/Territory	China
City	Shanghai
Period	9/04/14 → 10/04/14

UN SDGs

This output contributes to the following UN Sustainable Development Goals (SDGs)

SDG 3 Good Health and Well-being

Keywords

Geometric algorithm
Kernel methods
Scaled convex hull (SCH)
Support vector machines (SVMs)

Access to Document

10.4028/www.scientific.net/AMM.556-562.3935

Cite this

Guo, H. Y., Cheng, Y. Z., Wang, D. Z., & Xu, J. C. (2014). A complete and improved mathematical framework of the geometric approach based on the scaled convex hull (SCH) and its application to large medical data sets. In Mechatronics Engineering, Computing and Information Technology (pp. 3935-3940). (Applied Mechanics and Materials; Vol. 556-562). Trans Tech Publications. https://doi.org/10.4028/www.scientific.net/AMM.556-562.3935

@inproceedings{3018cc05fec446eabc78584f9402d1b6,

title = "A complete and improved mathematical framework of the geometric approach based on the scaled convex hull (SCH) and its application to large medical data sets",

abstract = "In this paper a complete and improved mathematical framework of the geometric approach based on the Scaled convex hull (SCH), which includes two parts, SCH-based geometric algorithms for SVM and a fast and novel geometric method for model selection in SCH-based SVM, is developed to solve SCH-based SVM classification of large data sets. On the basis of this framework, these geometric algorithms are more suitable for classification task of large data sets. Results of numerical experiments show that the proposed geometric algorithms can reduce kernel calculations and display nice performances on large medical data sets, such as computer assisted screening for lung cancer.",

keywords = "Geometric algorithm, Kernel methods, Scaled convex hull (SCH), Support vector machines (SVMs)",

author = "Guo, \{Hao Yan\} and Cheng, \{Yuan Zhi\} and Wang, \{Da Zheng\} and Xu, \{Jia Cheng\}",

year = "2014",

doi = "10.4028/www.scientific.net/AMM.556-562.3935",

language = "英语",

isbn = "9783038351153",

series = "Applied Mechanics and Materials",

publisher = "Trans Tech Publications",

pages = "3935--3940",

booktitle = "Mechatronics Engineering, Computing and Information Technology",

address = "德国",

note = "2014 International Conference on Mechatronics Engineering and Computing Technology, ICMECT 2014 ; Conference date: 09-04-2014 Through 10-04-2014",

}

Guo, HY, Cheng, YZ, Wang, DZ & Xu, JC 2014, A complete and improved mathematical framework of the geometric approach based on the scaled convex hull (SCH) and its application to large medical data sets. in Mechatronics Engineering, Computing and Information Technology. Applied Mechanics and Materials, vol. 556-562, Trans Tech Publications, pp. 3935-3940, 2014 International Conference on Mechatronics Engineering and Computing Technology, ICMECT 2014, Shanghai, China, 9/04/14. https://doi.org/10.4028/www.scientific.net/AMM.556-562.3935

A complete and improved mathematical framework of the geometric approach based on the scaled convex hull (SCH) and its application to large medical data sets. / Guo, Hao Yan; Cheng, Yuan Zhi; Wang, Da Zheng et al.
Mechatronics Engineering, Computing and Information Technology. Trans Tech Publications, 2014. p. 3935-3940 (Applied Mechanics and Materials; Vol. 556-562).

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

TY - GEN

T1 - A complete and improved mathematical framework of the geometric approach based on the scaled convex hull (SCH) and its application to large medical data sets

AU - Guo, Hao Yan

AU - Cheng, Yuan Zhi

AU - Wang, Da Zheng

AU - Xu, Jia Cheng

PY - 2014

Y1 - 2014

N2 - In this paper a complete and improved mathematical framework of the geometric approach based on the Scaled convex hull (SCH), which includes two parts, SCH-based geometric algorithms for SVM and a fast and novel geometric method for model selection in SCH-based SVM, is developed to solve SCH-based SVM classification of large data sets. On the basis of this framework, these geometric algorithms are more suitable for classification task of large data sets. Results of numerical experiments show that the proposed geometric algorithms can reduce kernel calculations and display nice performances on large medical data sets, such as computer assisted screening for lung cancer.

AB - In this paper a complete and improved mathematical framework of the geometric approach based on the Scaled convex hull (SCH), which includes two parts, SCH-based geometric algorithms for SVM and a fast and novel geometric method for model selection in SCH-based SVM, is developed to solve SCH-based SVM classification of large data sets. On the basis of this framework, these geometric algorithms are more suitable for classification task of large data sets. Results of numerical experiments show that the proposed geometric algorithms can reduce kernel calculations and display nice performances on large medical data sets, such as computer assisted screening for lung cancer.

KW - Geometric algorithm

KW - Kernel methods

KW - Scaled convex hull (SCH)

KW - Support vector machines (SVMs)

UR - https://www.scopus.com/pages/publications/84902094696

U2 - 10.4028/www.scientific.net/AMM.556-562.3935

DO - 10.4028/www.scientific.net/AMM.556-562.3935

M3 - 会议稿件

AN - SCOPUS:84902094696

SN - 9783038351153

T3 - Applied Mechanics and Materials

SP - 3935

EP - 3940

BT - Mechatronics Engineering, Computing and Information Technology

PB - Trans Tech Publications

T2 - 2014 International Conference on Mechatronics Engineering and Computing Technology, ICMECT 2014

Y2 - 9 April 2014 through 10 April 2014

ER -

Guo HY, Cheng YZ, Wang DZ, Xu JC. A complete and improved mathematical framework of the geometric approach based on the scaled convex hull (SCH) and its application to large medical data sets. In Mechatronics Engineering, Computing and Information Technology. Trans Tech Publications. 2014. p. 3935-3940. (Applied Mechanics and Materials). doi: 10.4028/www.scientific.net/AMM.556-562.3935

A complete and improved mathematical framework of the geometric approach based on the scaled convex hull (SCH) and its application to large medical data sets

Abstract

Publication series

Conference

UN SDGs

Keywords

Access to Document

Other files and links

Fingerprint

Cite this