CMAPP: Clustered matrix approximation 

(1) Decompress the cmapp.zip file
(2) Add the data folder into “MATLAB search paths” using “set path”
(3) Install GRACLUS, MatlabBGL, and optionally PROPACK (links below)
(4) Now you should be able to run all test files: 

cmappTest1.m
cmappTest2.m
cmappTest3.m
cmappTest4.m
cmappTest5.m
cmappTest6.m
cmappTest7.m

The clustered matrix approximation package uses: 

GRACLUS: Graph clustering software. 
The clustering of graphs and bipartite graphs in CMAPP is performed with GRACLUS. 
http://www.cs.utexas.edu/users/dml/Software/graclus.html

MatlabBGL: Matlab package for working with graphs. 
A number of files in CMAPP use a few files from MatlabBGL.
http://dgleich.github.io/matlab-bgl/

PROPACK: 
Software package for computing the singular value and spectral decomposition of large and sparse matrices. If available, it is possible to specify and used functions from PROPACK.
http://sun.stanford.edu/~rmunk/PROPACK/

Berkant Savas, 2015-05-03