weka - Clustering initialization -

how set clustering initialization method?

i found besides random initialization can select couple of more methods, such k-means++ , farthest first.

i found can use following method that:

clusterer.setinitializationmethod(new selectedtag);

now, i'm confused selectedtag. represent , how use it? more specifically, how specify k-means++ or farthest first initialization methods?

thanks

i found solution, here needs done:

clusterer.setinitializationmethod(new selectedtag(simplekmeans.kmeans_plus_plus, simplekmeans.tags_selection));

if @ simplekmeans see has following static members:

static int  canopy  static int  farthest_first  static int  kmeans_plus_plus  static int  random  static tag[]    tags_selection

and how use them. can pass distance identifier need.

cheers!

Soju