To improve the efficiency of standard PAM algorithm when deal with large data set, an improved fast PAM algorithm is put forward. The fast PAM algorithm combines the concept of spatial grid structure, by optimizing the choice of the initial representative objects and restricting the traverse object number in iterative process to reduce the computation work of standard PAM. Experimental results demonstrate the effectiveness of the fast PAM algorithm, which improve the performance and may save about 85% running time,while retaining exactly the same clustering quality comparing with the original standard PAM algorithm.%为提高标准PAM算法处理大数据集合的效率,提出了一种改进的快速PAM算法.该算法结合空间网格结构的概念,通过优化初始代表对象的选择、限制迭代过程中遍历的对象数量来残少标准队M算法的运算量.实验结果表明,相对于标准PAM算法,在保证聚类结果准确性的前提下,快速PAM算法可节省85%左右的执行时间,有效地改善了原算法的性能.
展开▼