Research on an Improved SVM Training Algorithm

Pan Feng; Danyang Qin; Ping Ji; Min Zhao; Ruolin Guo; Guangchao Xu; Lin Ma

doi:10.1007/978-981-13-9409-6_201

Research on an Improved SVM Training Algorithm

Pan Feng
, Danyang Qin^*
, Ping Ji
, Min Zhao
, Ruolin Guo
, Guangchao Xu
, Lin Ma

^*Corresponding author for this work

School of Electronics and Information Engineering

Heilongjiang University

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

Abstract

A new SVM training algorithm is proposed in the paper to improve the validity and efficiency of image annotation. These annotation tasks are related to one another due to the correlation among the labels. The model will implicitly learn a linear output kernel during training. Simulation results show that compared with independent SVMs training, Joint SVM improves classification accuracy and efficiency substantially.

Original language	English
Title of host publication	Communications, Signal Processing, and Systems - Proceedings of the 8th International Conference on Communications, Signal Processing, and Systems, CSPS 2019
Editors	Qilian Liang, Wei Wang, Xin Liu, Zhenyu Na, Min Jia, Baoju Zhang
Publisher	Springer
Pages	1674-1680
Number of pages	7
ISBN (Print)	9789811394089
DOIs	https://doi.org/10.1007/978-981-13-9409-6_201
State	Published - 2020
Event	8th International Conference on Communications, Signal Processing, and Systems, CSPS 2019 - Urumqi, China Duration: 20 Jul 2019 → 22 Jul 2019

Publication series

Name	Lecture Notes in Electrical Engineering
Volume	571 LNEE
ISSN (Print)	1876-1100
ISSN (Electronic)	1876-1119

Conference

Conference	8th International Conference on Communications, Signal Processing, and Systems, CSPS 2019
Country/Territory	China
City	Urumqi
Period	20/07/19 → 22/07/19

Keywords

Image annotation
Joint training
Output kernel
SVM

Access to Document

10.1007/978-981-13-9409-6_201

Cite this

Feng, P., Qin, D., Ji, P., Zhao, M., Guo, R., Xu, G., & Ma, L. (2020). Research on an Improved SVM Training Algorithm. In Q. Liang, W. Wang, X. Liu, Z. Na, M. Jia, & B. Zhang (Eds.), Communications, Signal Processing, and Systems - Proceedings of the 8th International Conference on Communications, Signal Processing, and Systems, CSPS 2019 (pp. 1674-1680). (Lecture Notes in Electrical Engineering; Vol. 571 LNEE). Springer. https://doi.org/10.1007/978-981-13-9409-6_201

Feng, Pan ; Qin, Danyang ; Ji, Ping et al. / Research on an Improved SVM Training Algorithm. Communications, Signal Processing, and Systems - Proceedings of the 8th International Conference on Communications, Signal Processing, and Systems, CSPS 2019. editor / Qilian Liang ; Wei Wang ; Xin Liu ; Zhenyu Na ; Min Jia ; Baoju Zhang. Springer, 2020. pp. 1674-1680 (Lecture Notes in Electrical Engineering).

@inproceedings{26387a7e9573448eb55f2c240dbd3c05,

title = "Research on an Improved SVM Training Algorithm",

abstract = "A new SVM training algorithm is proposed in the paper to improve the validity and efficiency of image annotation. These annotation tasks are related to one another due to the correlation among the labels. The model will implicitly learn a linear output kernel during training. Simulation results show that compared with independent SVMs training, Joint SVM improves classification accuracy and efficiency substantially.",

keywords = "Image annotation, Joint training, Output kernel, SVM",

author = "Pan Feng and Danyang Qin and Ping Ji and Min Zhao and Ruolin Guo and Guangchao Xu and Lin Ma",

note = "Publisher Copyright: {\textcopyright} 2020, Springer Nature Singapore Pte Ltd.; 8th International Conference on Communications, Signal Processing, and Systems, CSPS 2019 ; Conference date: 20-07-2019 Through 22-07-2019",

year = "2020",

doi = "10.1007/978-981-13-9409-6\_201",

language = "英语",

isbn = "9789811394089",

series = "Lecture Notes in Electrical Engineering",

publisher = "Springer",

pages = "1674--1680",

editor = "Qilian Liang and Wei Wang and Xin Liu and Zhenyu Na and Min Jia and Baoju Zhang",

booktitle = "Communications, Signal Processing, and Systems - Proceedings of the 8th International Conference on Communications, Signal Processing, and Systems, CSPS 2019",

address = "德国",

}

Feng, P, Qin, D, Ji, P, Zhao, M, Guo, R, Xu, G & Ma, L 2020, Research on an Improved SVM Training Algorithm. in Q Liang, W Wang, X Liu, Z Na, M Jia & B Zhang (eds), Communications, Signal Processing, and Systems - Proceedings of the 8th International Conference on Communications, Signal Processing, and Systems, CSPS 2019. Lecture Notes in Electrical Engineering, vol. 571 LNEE, Springer, pp. 1674-1680, 8th International Conference on Communications, Signal Processing, and Systems, CSPS 2019, Urumqi, China, 20/07/19. https://doi.org/10.1007/978-981-13-9409-6_201

Research on an Improved SVM Training Algorithm. / Feng, Pan; Qin, Danyang; Ji, Ping et al.
Communications, Signal Processing, and Systems - Proceedings of the 8th International Conference on Communications, Signal Processing, and Systems, CSPS 2019. ed. / Qilian Liang; Wei Wang; Xin Liu; Zhenyu Na; Min Jia; Baoju Zhang. Springer, 2020. p. 1674-1680 (Lecture Notes in Electrical Engineering; Vol. 571 LNEE).

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

TY - GEN

T1 - Research on an Improved SVM Training Algorithm

AU - Feng, Pan

AU - Qin, Danyang

AU - Ji, Ping

AU - Zhao, Min

AU - Guo, Ruolin

AU - Xu, Guangchao

AU - Ma, Lin

PY - 2020

Y1 - 2020

N2 - A new SVM training algorithm is proposed in the paper to improve the validity and efficiency of image annotation. These annotation tasks are related to one another due to the correlation among the labels. The model will implicitly learn a linear output kernel during training. Simulation results show that compared with independent SVMs training, Joint SVM improves classification accuracy and efficiency substantially.

AB - A new SVM training algorithm is proposed in the paper to improve the validity and efficiency of image annotation. These annotation tasks are related to one another due to the correlation among the labels. The model will implicitly learn a linear output kernel during training. Simulation results show that compared with independent SVMs training, Joint SVM improves classification accuracy and efficiency substantially.

KW - Image annotation

KW - Joint training

KW - Output kernel

KW - SVM

UR - https://www.scopus.com/pages/publications/85084748022

U2 - 10.1007/978-981-13-9409-6_201

DO - 10.1007/978-981-13-9409-6_201

M3 - 会议稿件

AN - SCOPUS:85084748022

SN - 9789811394089

T3 - Lecture Notes in Electrical Engineering

SP - 1674

EP - 1680

BT - Communications, Signal Processing, and Systems - Proceedings of the 8th International Conference on Communications, Signal Processing, and Systems, CSPS 2019

A2 - Liang, Qilian

A2 - Wang, Wei

A2 - Liu, Xin

A2 - Na, Zhenyu

A2 - Jia, Min

A2 - Zhang, Baoju

PB - Springer

T2 - 8th International Conference on Communications, Signal Processing, and Systems, CSPS 2019

Y2 - 20 July 2019 through 22 July 2019

ER -

Feng P, Qin D, Ji P, Zhao M, Guo R, Xu G et al. Research on an Improved SVM Training Algorithm. In Liang Q, Wang W, Liu X, Na Z, Jia M, Zhang B, editors, Communications, Signal Processing, and Systems - Proceedings of the 8th International Conference on Communications, Signal Processing, and Systems, CSPS 2019. Springer. 2020. p. 1674-1680. (Lecture Notes in Electrical Engineering). doi: 10.1007/978-981-13-9409-6_201

Research on an Improved SVM Training Algorithm

Abstract

Publication series

Conference

Keywords

Access to Document

Other files and links

Fingerprint

Cite this