P.S: This is for creating another thread of discussion

Hi Ryan

So I have been going through your Ph.D. thesis explaining the dual-tree
algorithms and how the BaseCase and Score are defined especially the
chapter-7(Algorithms).

So here is approach what I think for Fast-K center algorithm
1) Choose any arbitrary center and define a radius 'r' for that node. Now
for that node prune the neighbor nodes which are far by (alpha)*r (can be
same as borkuva algorithm pruning method)
2) The value of alpha can be iterated depending on the value of K.

Please let me know whether I'm going in right direction or not.

Ankit Aggarwal
_______________________________________________
mlpack mailing list
[email protected]
http://knife.lugatgt.org/cgi-bin/mailman/listinfo/mlpack

Reply via email to