摘要
提出了一种基于ai Net免疫网络模型的K-means聚类算法。该算法利用ai Net免疫网络模型中抗体-抗原之间的亲和力来计算聚类中心点,将数据分为若干子簇,之后再通过K-means聚类算法将这些子簇合并,得到最终的结果。该算法继承了免疫算法速度快,效率高的优点,同时也避免了K-means聚类算法容易陷入局部极小值的缺点,是一种高效的并行搜索算法。
This paper puts forward a K- means clustering algorithm based on ai Net immune network model. The algorithm calculates the clustering center point using the affinity between antibody and antigen in ai Net immune network model. The data is divided into several sub cluster and merge the clusters with the K- means clustering algorithm. The advantages of this algorithm is fast,high efficiency,and avoids the disadvantages of K- means clustering algorithm which is easy to fall into local minimum. It is an efficient parallel search algorithm.
出处
《延安大学学报(自然科学版)》
2015年第4期27-29,共3页
Journal of Yan'an University:Natural Science Edition