I am using K-means clustering algorithm, and I need to get clusters with one condition, that is, always each one includes one of the initial chosen centers, or if any other cluster methods which provide that feature.
update all points execpt the starting cluster points, as a markers for the clusters
Please check the following toolbox and its manual.
http://www.abonyilab.com/software-and-data/fclusttoolbox
Check this link.. is excellent.
http://en.wikipedia.org/wiki/K-means_clustering
Hi Tamer, what type of program you're using. send me an email I will send you a script k-means
now i am using c++ to execute k-mean algorithm, but i need an update for the k-means algorithm such that certain k-points to become fixed each in one different cluster, point i is a member of cluster j even after any iteration, or at least a condition to make that granted
This might not be a good idea to mark certain points before running the K-means algorithm as cluster representatives, since some points might end up being in the same cluster.
You can run the K-means algorithms on your data points and search for those "special" points in the clusters. Then you can mark those clusters as contatining those "special" data points.
in my problem those points are given and conditioned to be the cluster representative
go with this links
http://mnemstudio.org/clustering-k-means-example-1.htm
http://webdocs.cs.ualberta.ca/~zaiane/courses/cmput695/F07/exercises/Exercises695Clus-solution.pdf
in the last example the 1st centroids are point 1 for cluster one and point 4 for cluster two and finished in the same clusters, but that is not guaranteed in all cases, what i have to do to grandate that in all cases
What I need only that the 1st centroids set stay in their clusters
- Badges
- Science topic