This is the Matlab programs for KAP - generating a given number of clusters by AP.

Created by 
Xiangliang ZHANG

CopyRight (c) 2010-2012, Xiangliang ZHANG
All rights reserved.

%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
KAP is a clustering method generating a given number of clusters based on message-passing.

It was first proposed in the paper:

	Xiangliang Zhang, Wei Wang, Kjetil Nørvåg, Michèle Sebag, "K-AP: Generating Specified K Clusters by Efficient Affinity Propagation", in the proceedings of 10th IEEE International Conference on Data Mining (ICDM' 2010), Sydney, Australia, December 14-17, 2010

For having the newest version of KAP Matlab codes, welcome to my homepage
http://www.lri.fr/~xlzhang

%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
Usage (example):
in Matlab, type: clustering_toy

%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
Files:

cluster_toy.m      the main program used to produce K clusters
KAP.m              the clustering method KAP
apclusterK_frey.m  re-launch AP several times for getting K clusters
apcluster.m        clustering method AP
dis_cmp.m          computing the distortion of clustering
similarity_nse.m   used to obtain the negative squared error(Euclidean distance) for each pair of examples.
ToyProblem.mat     the toy data

%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
If you used this software, please cite it as

@inproceedings{xlzhang_ICDM10,
  author = {Xiangliang Zhang and Wei Wang and Kjetil Norvag and Michele Sebag},
  title = {{K-AP: Generating Specified K Clusters by Efficient Affinity Propagation}},
  booktitle = {IEEE International Conference on Data Mining (ICDM)},
  pages = {1187-1192},
  year = 2010,
}


%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
If you have any questions or comments, please send xlzhang.email@gmail.com

For more information and newer version, please visit http://www.lri.fr/~xlzhang
Enjoy the software!

2010-12-23