Ask Your Question
0

hierarchical Clustering VS Kmeans Clustering

asked 2013-07-31 01:05:43 -0600

Mostafa Sataki gravatar image Mostafa Sataki flag of Iran, Islamic Republic of
1793 3 8 26
http://www.7khatcode.com/

updated 2013-07-31 06:42:10 -0600

StevenPuttemans gravatar image StevenPuttemans flag of Belgium
9004 3 30 89
http://stevenputtemans.gi...

What is difference between hierarchical Clustering & Kmeans clustering?What is advantage of hierarchical Clustering to Kmeans clustering?

delete close flag offensive retag edit

1 Answer

Sort by ยป oldest newest most voted
1

answered 2013-07-31 02:41:41 -0600

red_sonya gravatar image red_sonya
11

updated 2013-07-31 02:53:28 -0600

Siegfried gravatar image Siegfried
988 3 13

That is not bad the explanation of the differences between hierarchical and k-means clustering(page 46): http://www.lcb.uu.se/KBIB/materials/L1.pdf. And there is an rational thought about the using of hierarchical k-means (top-down hierarchical clustering using k-means iteratively) there. The advantage of one or the other method depends of the task's conditions.

link delete flag offensive edit

Comments

One of the differences between these two algorithms is number of center.In the Kmeans have to determine K param but in hierarchical clustering does not require specify K param & algorithm determines K param.Why in the opencv hierarchical Clustering have to send center Mat?

Mostafa Sataki ( 2013-07-31 03:29:40 -0600 )edit

If you mean flann::hierarchicalClustering in OpenCV it is a hierarchical k-means tree. You need to create a matrix for centers. The centers.rows will be the upper limit of clusters which you need. This matrix will contain the obtained centers of clusters.

red_sonya ( 2013-07-31 05:03:20 -0600 )edit

Login/Signup to Answer

Question tools

Follow

subscribe to rss feed

Stats

Asked: 2013-07-31 01:05:43 -0600

Seen: 2,487 times

Last updated: Jul 31 '13