动态遗传算法在大数据聚类分析中的应用

A Dynamic Clustering Method Based on Genetic Algorithms For BIG Data Analysis

对于常规聚类方法,聚类结果往往与初始聚类中心数目和数据入次序有关。本文另辟蹊径,提出了一种动态GA来实现样本类别数目由数据本身来确定,避免了聚类数目确定的盲目性等问题。计算结果表明,该方法是一个具有全局最优解的动态聚类方法。

To solve the problem of sensitivity with original clustering center;clustering results depended on the order of inputted example in common dynamic clustering algorithm,and blindness of decided the number of clusters,a new dynamic clustering method based on genetic algorithms is presented in this paper and is applied for data auto clustering in databases.Computed results indicate that the method is a dynamic clustering algorithm with global optimization.