A multi-objective particle swarm optimization algorithm for rule discovery

Sheng Tun Li, Chih Chuan Chen, Jian Wei Li

研究成果: Conference contribution

4 引文 (Scopus)

摘要

Rule discovery is usually posed as a multi-objective optimization problem with two criteria, predictive accuracy and comprehensibility. Single-objective particle swarm optimization algorithm, which combines the two criteria into one, has been shown to have convincing results on the classification tasks. However, it does not take the nature of the optimality conditions for multiple objectives into account. It is well known that accuracy and comprehensibility are hardly attainable simultaneously, which makes the optimization problem difficult to solve efficiently. In this paper, we propose a multi-objective PSO algorithm to solve the problem. The experimental result shows that our algorithm has better performance than its single-objective counterpart.

原文English
主出版物標題Proceedings - 3rd International Conference on Intelligent Information Hiding and Multimedia Signal Processing, IIHMSP 2007.
頁面597-600
頁數4
DOIs
出版狀態Published - 2007 十二月 1
事件3rd International Conference on Intelligent Information Hiding and Multimedia Signal Processing, IIHMSP 2007 - Kaohsiung, Taiwan
持續時間: 2007 十一月 262007 十一月 28

出版系列

名字Proceedings - 3rd International Conference on Intelligent Information Hiding and Multimedia Signal Processing, IIHMSP 2007.
2

Other

Other3rd International Conference on Intelligent Information Hiding and Multimedia Signal Processing, IIHMSP 2007
國家Taiwan
城市Kaohsiung
期間07-11-2607-11-28

指紋

Particle swarm optimization (PSO)
Multiobjective optimization
Particle swarm optimization
Optimization problem
Multiple objectives
Predictive accuracy
Multi-objective optimization
Optimality conditions

All Science Journal Classification (ASJC) codes

  • Computer Networks and Communications
  • Signal Processing
  • Information Systems and Management

引用此文

Li, S. T., Chen, C. C., & Li, J. W. (2007). A multi-objective particle swarm optimization algorithm for rule discovery. 於 Proceedings - 3rd International Conference on Intelligent Information Hiding and Multimedia Signal Processing, IIHMSP 2007. (頁 597-600). [4457780] (Proceedings - 3rd International Conference on Intelligent Information Hiding and Multimedia Signal Processing, IIHMSP 2007.; 卷 2). https://doi.org/10.1109/IIH-MSP.2007.34
Li, Sheng Tun ; Chen, Chih Chuan ; Li, Jian Wei. / A multi-objective particle swarm optimization algorithm for rule discovery. Proceedings - 3rd International Conference on Intelligent Information Hiding and Multimedia Signal Processing, IIHMSP 2007.. 2007. 頁 597-600 (Proceedings - 3rd International Conference on Intelligent Information Hiding and Multimedia Signal Processing, IIHMSP 2007.).
@inproceedings{76ea4896b97b425da509ed1ad668029f,
title = "A multi-objective particle swarm optimization algorithm for rule discovery",
abstract = "Rule discovery is usually posed as a multi-objective optimization problem with two criteria, predictive accuracy and comprehensibility. Single-objective particle swarm optimization algorithm, which combines the two criteria into one, has been shown to have convincing results on the classification tasks. However, it does not take the nature of the optimality conditions for multiple objectives into account. It is well known that accuracy and comprehensibility are hardly attainable simultaneously, which makes the optimization problem difficult to solve efficiently. In this paper, we propose a multi-objective PSO algorithm to solve the problem. The experimental result shows that our algorithm has better performance than its single-objective counterpart.",
author = "Li, {Sheng Tun} and Chen, {Chih Chuan} and Li, {Jian Wei}",
year = "2007",
month = "12",
day = "1",
doi = "10.1109/IIH-MSP.2007.34",
language = "English",
isbn = "0769529941",
series = "Proceedings - 3rd International Conference on Intelligent Information Hiding and Multimedia Signal Processing, IIHMSP 2007.",
pages = "597--600",
booktitle = "Proceedings - 3rd International Conference on Intelligent Information Hiding and Multimedia Signal Processing, IIHMSP 2007.",

}

Li, ST, Chen, CC & Li, JW 2007, A multi-objective particle swarm optimization algorithm for rule discovery. 於 Proceedings - 3rd International Conference on Intelligent Information Hiding and Multimedia Signal Processing, IIHMSP 2007.., 4457780, Proceedings - 3rd International Conference on Intelligent Information Hiding and Multimedia Signal Processing, IIHMSP 2007., 卷 2, 頁 597-600, 3rd International Conference on Intelligent Information Hiding and Multimedia Signal Processing, IIHMSP 2007, Kaohsiung, Taiwan, 07-11-26. https://doi.org/10.1109/IIH-MSP.2007.34

A multi-objective particle swarm optimization algorithm for rule discovery. / Li, Sheng Tun; Chen, Chih Chuan; Li, Jian Wei.

Proceedings - 3rd International Conference on Intelligent Information Hiding and Multimedia Signal Processing, IIHMSP 2007.. 2007. p. 597-600 4457780 (Proceedings - 3rd International Conference on Intelligent Information Hiding and Multimedia Signal Processing, IIHMSP 2007.; 卷 2).

研究成果: Conference contribution

TY - GEN

T1 - A multi-objective particle swarm optimization algorithm for rule discovery

AU - Li, Sheng Tun

AU - Chen, Chih Chuan

AU - Li, Jian Wei

PY - 2007/12/1

Y1 - 2007/12/1

N2 - Rule discovery is usually posed as a multi-objective optimization problem with two criteria, predictive accuracy and comprehensibility. Single-objective particle swarm optimization algorithm, which combines the two criteria into one, has been shown to have convincing results on the classification tasks. However, it does not take the nature of the optimality conditions for multiple objectives into account. It is well known that accuracy and comprehensibility are hardly attainable simultaneously, which makes the optimization problem difficult to solve efficiently. In this paper, we propose a multi-objective PSO algorithm to solve the problem. The experimental result shows that our algorithm has better performance than its single-objective counterpart.

AB - Rule discovery is usually posed as a multi-objective optimization problem with two criteria, predictive accuracy and comprehensibility. Single-objective particle swarm optimization algorithm, which combines the two criteria into one, has been shown to have convincing results on the classification tasks. However, it does not take the nature of the optimality conditions for multiple objectives into account. It is well known that accuracy and comprehensibility are hardly attainable simultaneously, which makes the optimization problem difficult to solve efficiently. In this paper, we propose a multi-objective PSO algorithm to solve the problem. The experimental result shows that our algorithm has better performance than its single-objective counterpart.

UR - http://www.scopus.com/inward/record.url?scp=47349090621&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=47349090621&partnerID=8YFLogxK

U2 - 10.1109/IIH-MSP.2007.34

DO - 10.1109/IIH-MSP.2007.34

M3 - Conference contribution

AN - SCOPUS:47349090621

SN - 0769529941

SN - 9780769529943

T3 - Proceedings - 3rd International Conference on Intelligent Information Hiding and Multimedia Signal Processing, IIHMSP 2007.

SP - 597

EP - 600

BT - Proceedings - 3rd International Conference on Intelligent Information Hiding and Multimedia Signal Processing, IIHMSP 2007.

ER -

Li ST, Chen CC, Li JW. A multi-objective particle swarm optimization algorithm for rule discovery. 於 Proceedings - 3rd International Conference on Intelligent Information Hiding and Multimedia Signal Processing, IIHMSP 2007.. 2007. p. 597-600. 4457780. (Proceedings - 3rd International Conference on Intelligent Information Hiding and Multimedia Signal Processing, IIHMSP 2007.). https://doi.org/10.1109/IIH-MSP.2007.34