Code covered by the BSD License  

Highlights from
Fast and efficient spectral clustering

image thumbnail

Fast and efficient spectral clustering


Ingo (view profile)


02 Jan 2012 (Updated )

Perform fast and efficient spectral clustering algorithms

SimGraph_Full(M, sigma)
function W = SimGraph_Full(M, sigma)
% SIMGRAPH_FULL Returns full similarity graph
%   Returns adjacency matrix for a full similarity graph where
%   a Gaussian similarity function with parameter sigma is
%   applied.
%   'M' - A d-by-n matrix containing n d-dimensional data points
%   'sigma' - Parameter for Gaussian similarity function
%   Author: Ingo Buerk
%   Year  : 2011/2012
%   Bachelor Thesis

% Compute distance matrix
W = squareform(pdist(M'));

% Apply Gaussian similarity function
W = simGaussian(W, sigma);


Contact us