• Deng Cai, Xiaofei He, and Jiawei Han, "Document Clustering Using Locality Preserving Indexing", in IEEE TKDE, 2005.
  Bibtex source
• Xinlei Chen, Deng Cai, "Large Scale Spectral Clustering with Landmark-Based Representation," AAAI 2011.
  Bibtex source

Algorithms

• Please see here for clustering using Graph regularized Gaussian Mixture Model.

  ---

• Clustering Evaluation: Evaluate the clustering result by accuracy and normalized mutual information
  • Deng Cai, Xiaofei He, and Jiawei Han, "Document Clustering Using Locality Preserving Indexing", in IEEE TKDE, 2005.
    Bibtex source

  • bestMap
  • hungarian
  • MutualInfo

    fea = rand(50,70);
    gnd = [ones(10,1);ones(15,1)*2;ones(10,1)*3;ones(15,1)*4];
    
    res = kmeans(fea,4);
    
    res = bestMap(gnd,res);
    %=============  evaluate AC: accuracy ==============
    AC = length(find(gnd == res))/length(gnd);
    %=============  evaluate MIhat: nomalized mutual information =================
    MIhat = MutualInfo(gnd,res);
    

  ---

• litekmeans: The fastest matlab implementation of kmeans.
  • Bibtex source

    % Windows 7 on Intel(R) Core(TM) i7-2600K CPU @ 3.40GHz with 16G memory 
    % Matlab R2011b (7.13.0.564) 64-bit (win64) 
    
    % MNIST data set (70000 samples with 784 dimensions) 
    clear;
    load('Orig.mat'); 
    
    % litekmeans
    tic;[label, center] = litekmeans(fea, 200, 'MaxIter', 20);toc
    Elapsed time is 11.515731 seconds. 
    
    % kmeans provided in Matlab
    tic;[label, center] = kmeans(fea, 200, 'OnlinePhase','off','MaxIter', 20);toc
    Elapsed time is 237.617422 seconds.
    

  ---

• LSC: Landmark-based Spectral Clustering
  • litekmeans, mySVD and EuDist2 are required.
  • Example here

  • Xinlei Chen, Deng Cai, "Large Scale Spectral Clustering with Landmark-Based Representation," AAAI 2011. (pdf)
    Bibtex source