数据分布密度划分的聚类算法是数据挖掘聚类算法的主要方法之一。针对传统密度划分聚类算法存在运算复杂、运行效率不高等缺陷,设计高维分步投影的多重分区聚类算法;以高维分布投影密度为依据,对数据集进行多重分区,产生数据集的子簇空间,并进行子簇合并,形成理想的聚类结果;依据该算法进行实验,结果证明该算法具有运算简单和运行效率高等优良性。%Clustering algorithm based on the data distribution density division is one of the main methods of data mining clustering algorithm.In view of the complex algorithm and low operation efficiency of traditional density-classification clustering algorithm,the authors of this paper design a multiple-partition clustering algorithm of high dimensional step proj ection. On account of high dimensional distribution proj ection density,we apply multi-partitonig of data sets to generate the sub cluster space from data sets,and merge sub clusters to form desired clustering results.Experiments'results show that this algorithm perform excellently in operation and efficiency.
展开▼