How to evaluate the performance of K-means?

1 次查看（过去 30 天）
Salad Box
Salad Box 2019-12-4
编辑： Adam Danz 2019-12-4
Hi,
I would like to evaluate the performance of k-means. I saw the below cost function (1) online, where wik=1 for data point xi if it belongs to cluster k; otherwise, wik=0. Also, μk is the centroid of xi’s cluster. But I don't know what 'm' is.
Does anyone understand what 'm' is in the below cost function?

请先登录，再回答此问题。

采纳的回答

Adam Danz
Adam Danz 2019-12-4
编辑：Adam Danz 2019-12-4
"...where wik=1 for data point xi..."
That would suggest that m is the number of data points since that summation is iterating over i.
Indeed, this is the correct interpretation (source).

更多回答（0 个）

类别

在 Help Center 和 File Exchange 中查找有关 k-Means and k-Medoids Clustering 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