Clustering method with evolutionary immune network based on polyclonal algorithm

doi:10.3778/j.issn.1002-8331.2009.27.044

Computer Engineering and Applications ›› 2009, Vol. 45 ›› Issue (27): 146-150.DOI: 10.3778/j.issn.1002-8331.2009.27.044

• 数据库、信号与信息处理 • Previous Articles Next Articles

Clustering method with evolutionary immune network based on polyclonal algorithm

ZHOU Yang¹，MA Li²，BAI Lin²

1.Department of Computer Science and Technology，Xi’an Institute of Post and Telecommunications，Xi’an 710061，China
2.Information Center，Xi’an Institute of Post and Telecommunications，Xi’an 710061，China

Received:2008-05-26 Revised:2008-08-18 Online:2009-09-21 Published:2009-09-21
Contact: ZHOU Yang

基于多克隆的进化免疫网络聚类算法

周洋¹，马力²，白琳²

1.西安邮电学院计算机科学与技术，西安 710061
2.西安邮电学院信息中心，西安 710061

通讯作者: 周洋

Abstract

Abstract: The traditional algorithms of cluster are very sensitive to the initialization and easy to get trapped into local optima，Moreover these algorithms depend more on prior knowledge about the cluster number and the type of clustering prototypes.For this purpose，a clustering method with evolutionary immune network based on polyclonal algorithm is presented，this algorithm is developed by improving clonal selection algorithm based on immune network.The algorithm uses polyclonal operator，just because of this operation，the diversity of antibody group is broadened and the search scope of solution space is widened.The algorithm uses forbidden clone operation so that the antibodies at the fuzzy boundary is in the state of suppression，this operation improves the accuracy of the cluster.The experiment shows that the algorithm has fast convergence and is robust to initialization while clustering for the data sets which have not only mixed numerical and categorical values but also fuzzy boundary values.

Key words: polyclonal algorithm, immune network, clonal selection, forbidden clone

摘要： 针对传统的聚类算法存在对初始值敏感、易陷入局部最小值，且对类别数和聚类原型的先验知识依赖比较大等问题。提出了一种基于多克隆的进化免疫网络聚类算法，该算法使用了多克隆算子，增加了种群的多样性，扩大了解空间的搜索范围。利用禁忌克隆运算，使处于模糊边界的抗体处于抑制状态，提高了聚类的精度。仿真实验表明，当对具有数值和类属的混合特征属性的数据及具有模糊边界的数据进行聚类时，收敛速度快且不依赖初始原型的选择。

关键词: 多克隆算法, 免疫网络, 克隆选择, 禁忌克隆

CLC Number:

TP391

ZHOU Yang¹，MA Li²，BAI Lin². Clustering method with evolutionary immune network based on polyclonal algorithm[J]. Computer Engineering and Applications, 2009, 45(27): 146-150.

周洋¹，马力²，白琳². 基于多克隆的进化免疫网络聚类算法[J]. 计算机工程与应用, 2009, 45(27): 146-150.

[1]	CHEN Haiyang, JIN Xiaolei, NIU Longhui, LIU Xiqing. Optimization of Regional Traffic Lights Timing Based on Improved Clonal Selection Algorithm [J]. Computer Engineering and Applications, 2020, 56(9): 272-278.
[2]	LIU Ximei, PAN Lijun. Clonal Selection Algorithm for Solving Bike Sharing Rebalancing Problem [J]. Computer Engineering and Applications, 2020, 56(11): 252-258.
[3]	WU Dongkai, ZHANG Fengbin, XI Liang. Research on clonal selection algorithm of adaptive hybrid mutation [J]. Computer Engineering and Applications, 2018, 54(21): 78-83.
[4]	ZHANG Fengbin, WANG Chao, YANG Ze. Clonal selection algorithm based on vaccination and Cauchy mutation [J]. Computer Engineering and Applications, 2017, 53(1): 118-123.
[5]	DONG Feng1, ZHOU Pengxu2. Multi-level immune intrusion detection model for cloud computing [J]. Computer Engineering and Applications, 2016, 52(21): 101-104.
[6]	LIU Hanqiang. Semisupervised immune clone selection graph partition algorithm [J]. Computer Engineering and Applications, 2014, 50(22): 11-16.
[7]	JIA Huaping1, LI Yaolong1, SHI Xiaoying2. Network intrusion detection model of chaos immune genetic algorithm [J]. Computer Engineering and Applications, 2014, 50(21): 96-99.
[8]	XU Chengqian, HAO Hongjie. Improved intelligent optimization algorithm used in multiuser detection [J]. Computer Engineering and Applications, 2013, 49(3): 143-146.
[9]	TONG Chengyi. Image threshold segmentation method based on CSACS [J]. Computer Engineering and Applications, 2013, 49(24): 161-164.
[10]	YANG Gelan1, JIN Huixia2, ZHU Xinghui3. Optimization algorithm based on differential evolution and clonal selection mechanism [J]. Computer Engineering and Applications, 2013, 49(10): 50-52.
[11]	PAN Xiaoying, CHEN Hao. License plate location based on immune clonal selection [J]. Computer Engineering and Applications, 2012, 48(3): 162-164.
[12]	LAN Fangpeng1, DUAN Fu2. Immune optimization algorithm for shipping arrangement of water transport of coal [J]. Computer Engineering and Applications, 2012, 48(16): 224-228.
[13]	LIU Qian1, QIU Bin2. Flowers image segmentation based on clonal selection algorithm [J]. Computer Engineering and Applications, 2012, 48(14): 185-189.
[14]	PENG Libing，HUANG Huixian，RUAN Ting，YI Liu. Adaptive niche clonal selection algorithm for multi-modal function optimization [J]. Computer Engineering and Applications, 2011, 47(9): 48-50.
[15]	LI Yang1，YUE Xiaobo2. Research on clonal selection algorithm applied in parameters optimization of fuzzy Petri nets [J]. Computer Engineering and Applications, 2011, 47(15): 39-42.

Clustering method with evolutionary immune network based on polyclonal algorithm

基于多克隆的进化免疫网络聚类算法

PDF

Knowledge

Abstract

Cite this article

share this article

References

Related Articles 15

Recommended Articles

Metrics