Kmeans information and cluster analysis

1 次查看（过去 30 天）
George Leavitt
George Leavitt 2016-11-28
回答： Nirja Mehta 2016-12-1
Hello so I recently used kmeans clustering to group customers with similar purchase order histories/customer data. I have obtained the IDk, c , sumd, and d values obtained from the kmeans function. My question is with this data, How am I supposed to extract information so I can learn about what is in the clusters? Are there any functions or methods to describe the affects of the variables? Please help.

请先登录，再回答此问题。

回答（1 个）

Nirja Mehta
Nirja Mehta 2016-12-1
The outputs of kmeans will help you in visualizing the way your data is separated among k number of clusters. The outputs contain the following information:
idx - cluster indices
C — Cluster centroid locations
sumd — Within-cluster sums of point-to-centroid distances
D — Distances from each point to every centroid
With this information, you can plot the clusters and the cluster centroids as shown in the following link: K-Means Plot

类别

在 Help Center 和 File Exchange 中查找有关 Statistics and Machine Learning Toolbox 的更多信息

标签

Community Treasure Hunt

Find the treasures in MATLAB Central and discover how the community can help you!

Start Hunting!

Translated by