关键词: 人工免疫系统 核聚类人工免疫网络 超变异率 连接阈值

Abstract:

The Kernel Clustering Artificial Immune Network is a new clustering analysis method,its training algorithm is also a kind of stochastic searching algo rithm,so the performance of the algorithm is dependent on several parameters.This paper discusses the effect of these key parameters on the algorithm by   experimental results,and then suggests how to select these parameters properly.

Key words: artificial immune system;kernel clustering artificial immune network;hypermutation rate;connect threshold