基于矩阵变换的聚类集成优化模型

聚类集成方法能够有效综合不同的聚类结果,提高聚类的精确度和稳定性.提出了一个基于矩阵变换的聚类集成优化模型,模型通过矩阵变换代替传统方法中的聚类配准模式,使得优化模型更加简洁,然后给出了求解该优化模型的叠代算法.实验表明,提出的聚类集成方法能够有效提高聚类集成的稳定性和精确度,并且在聚类数目比较少时,算法有着较低的时间复杂度.

An Optimization Model about Clustering Ensemble based on Matrix Transformation

The accuracy and stability of clustering could be enhanced by integrating different clustering results with clustering ensemble methods.In this study,an optimization model for clustering ensemble is presented based on matrix transformation,which is used to replace the traditional clustering matching mode and make the optimization model concise. Then an iterative algorithm is brought out.The demonstration indicates that this optimization model can effectively improve the accuracy and stability of clustering,and the iterative algorithm has lower computational complexity while the number of clusters is not large.