摘要
UK均值算法需要计算每个对象之间的期望距离(EDS)和聚类中心,EDS计算的成本就成了UK均值计算的性能瓶颈。为了提高UK均值的计算效率,本文提出一种优化的UK均值算法,通过一个高效的公式来估计期望距离,大大降低了UK均值的额外时间,并在实验中得以证明。我们还说明这个优化公式有效地将UK均值算法降低到了传统的基于K均值的聚类算法。
UK-means algorithm needs to calculate the desired distance between each object (EDS) and duster centers, EDS calculated the cost of performance bottlenecks became the UK mean calculation. In order to improve the computational efficiency of the UK mean the UK mean an optimization algorithm, an efficient formula to estimate the expected distance, greatly reducing the UK mean extra time and in evidenced experiment. We also show that this optimized formula effectively reduced to the UK-means algorithm based K-means clustering algorithm.
出处
《计算技术与自动化》
2013年第2期60-63,共4页
Computing Technology and Automation
关键词
不确定数据
聚类
期望距离
UK均值算法
.uncertain data
clustering
expected distance
the UK average value algorithm