摘要
给出了一种基于人工免疫的可更新簇聚类算法。该算法在aiNet聚类算法的基础上,引入记忆抗体"年龄"的概念。模型学习过程中在抗体不断地与抗原接触时,超过"年龄"阈值的记忆抗体转化为一般抗体,以此达到聚类簇的更新。实验表明该算法是可行、有效的。
In this paper,an updated cluster algorithm based on artificial immune is proposed,This algorithm is proposed based on the algorithm of aiNet cluster,and a new conception of ＂age＂ of the memory antibody is introduced.During the learning process, when the ＂age＂ of memory antibody exceeds the threshold of the memory＂ antibody,the memory antibody will change into common antibody.It＇s very feasible and effective through experiment.
出处
《计算机工程与应用》
CSCD
北大核心
2007年第27期168-170,180,共4页
Computer Engineering and Applications
