How to evaluate the performance of K-means?

1 visualización (últimos 30 días)
Salad Box
Salad Box el 4 de Dic. de 2019
Editada: Adam Danz el 4 de Dic. de 2019
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?

Iniciar sesión para responder a esta pregunta.

Respuesta aceptada

Adam Danz
Adam Danz el 4 de Dic. de 2019
Editada: Adam Danz el 4 de Dic. de 2019
"...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).

Más respuestas (0)

Categorías

Más información sobre k-Means and k-Medoids Clustering en Help Center y File Exchange.

Etiquetas

Community Treasure Hunt

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

Start Hunting!

Translated by