Ckmeans.1d.dp: Optimal k-means Clustering in One Dimension by Dynamic Programming
Ckmeans.1d.dp: Optimal k-means Clustering in One Dimension by Dynamic Programming
复制标题
DOI：
10.32614/rj-2011-015
复制
发表时间：
2011-12-01
期刊：
影响因子：
2.1
通讯作者：
Song, Mingzhou
中科院分区：
文献类型：
--
作者：
Wang, Haizhou;Song, Mingzhou
启发式k-means算法，广泛用于聚类分析，不保证最优。我们开发了一个动态规划算法的最佳一维聚类。该算法被实现为一个名为Ckmeans.1d.dp的R包。我们证明了它的优势，在最优性和运行时间超过标准的迭代k-means算法。
The heuristic k-means algorithm, widely used for cluster analysis, does not guarantee optimality. We developed a dynamic programming algorithm for optimal one-dimensional clustering. The algorithm is implemented as an R package called Ckmeans.1d.dp. We demonstrate its advantage in optimality and runtime over the standard iterative k-means algorithm.