source: distools/sphem.m @ 13

%SPHEM Spherical Embedding
%
%       [W,SIG,L] = SPHEM(D,R,ALF)
%                 OR
%       [W,SIG,L] = SPHEM(W,ALF)
%
% INPUT
%   D   NxN symmetric dissimilarity matrix (dataset)
%   W   Trained linear embedding into a pseudo-Euclidean space
%   ALF Parameter determining the dimensionality and the mapping (optional, defaulf: Inf)
%       (0,1)   - fraction of the total (absolute value) preserved variance
%        Inf    - no dimensionality reduction, keeping all dimensions (it's noisy)
%       'p'     - projection into a Euclidean space based on positive eigenvalues only
%       'PARp'  - projection into a Euclidean space based on the PAR fraction of
%                 positive eigenvalues; e.g. ALF = '0.9p'
%       'n'     - projection into a Euclidean space based on negative eigenvalues only
%       'PARn'  - projection into a (negative) Euclidean space based on the PAR fraction
%                 of negative eigenvalues; e.g. ALF = '0.7n'
%       'P1pP2n'- projection into a Euclidean space based on the P1 positive eigenvalues
%                 and P2 negative eigenvalues; e.g. ALF = '0.7p0.1n', ALF = '7p2n'
%       1 .. N  - number of dimensions in total
%       [P1 P2] - P1 dimensions or preserved fraction of variance in the positive subspace
%                 and P2 dimensions or preserved fraction of variance in the negative
%                 subspace; e.g. ALF = [5 10], ALF = [0.9 0.1]
%   P   Integer between 0 and N specifying which object is mapped at the origin;
%       0 stands for the mean; (optional, default: 0)
%
% OUTPUT
%   W   Linear embedding into a pseudo-Euclidean space
%   SIG Signature of the space
%   L   List of eigenvalues
%
% DESCRIPTION
% Linear mapping W onto an M-dimensional pseudo-Euclidean subspace from a symmetric,
% square dissimilarity matrix D such that the dissimilarities are preserved. M
% M is determined by ALF. E.g., the subspace is found such that at least a fraction
% ALF of the total variance is preserved for ALF in (0,1). The resulting X is found
% by D*W. The parameter SIG describes the signature of the subspace. L is a sorted
% list of eigenvalues describing the variances in the (pseudo-)Euclidean space.
%
% A trained mapping can be reduced further by:
%   [W,SIG,L] = SPHEM(D,ALF)
%
% SEE ALSO
% MAPPINGS, DATASETS, PSEM, AUGPSEM, PCA
%
% LITERATURE
% 1. L. Goldfarb, A unified approach to pattern recognition, Pattern Recognition, vol.17, 575-582, 1984.
% 2. E. Pekalska, P. Paclik, and R.P.W. Duin, A Generalized Kernel Approach to Dissimilarity-based
%    Classification, Journal of Machine Learning Research, vol.2, no.2, 175-211, 2002.
%

% Copyright: Elzbieta Pekalska, ela.pekalska@googlemail.com
% Faculty EWI, Delft University of Technology and
% School of Computer Science, University of Manchester



function [W,sig,L,Q] = sphem(d,R,alf,prec)
if nargin < 4 | isempty(prec), prec = 1e-4; end
if nargin < 3 | isempty(alf), alf = inf; end
if nargin < 2 | isempty(R), R = 1; end
if nargin < 1 | isempty(d),
  W = mapping(mfilename,{R,alf,prec});
  W = setname(W,'Spherical embedding');
  return
end



if (isdataset(d) | isa(d,'double'))
  if ismapping(R)

    % APPLY THE MAPPING
    pars  = getdata(R);
    [m,n] = size(d);

    R = pars{1};  % radius
    Q = pars{2};  % Eigenvectors
    L = pars{3};  % Eigenvalues

    W = R^2*(cos(+d/R)) * Q * diag(sqrt(abs(L))./L);
    if isdataset(d),
      W = dataset(W,getlab(d),'prior',getprior(d));

      % Store signature in the USER field
      W.user = pars{4};    % Signature
      W.name = ['Projected '  updname(d.name)];
    end
    return
  end
end



% REDUCE ALREADY TRAINED MAPPING
if ismapping(d)
  pars = getdata(d);
  R = pars{1};  % radius
  Q = pars{2};  % Eigenvectors
  L = pars{3};
  m = size(Q,1);

  [ll,K] = sort(-abs(L));
  L = L(K);
  Q = Q(:,K);
  [J,sig] = seleigs(L,alf,pars{5});
  Q = Q(:,J);        % Eigenvectors
  L = L(J);          % Eigenvalues

  W = mapping(mfilename,'trained',{R,Q,L,pars{4},pars{5}},[],m,length(J));
  W = setname(W,'Spherical embedding');
  return
end



% TRAIN THE MAPPING

% Tolerance value used in comparisons
if mean(+d(:)) < 1,
  tol = 1e-12;
else
  tol = 1e-10;
end

if R <= 0,
  error('R should be positive.');
end


[n,m] = size(d);
if ~issym(d,tol),
  prwarning(1,'Matrix should be symmetric. It is made symmetric by averaging.')
  d = 0.5*(d+d');
end

B = R^2*cos(+d/R);
B = 0.5*(B+B');                       % Make sure B is symmetric

[Q, L] = preig(B);
Q      = real(Q);
l      = diag(real(L));
[lm,Z] = sort(-abs(l));
Q      = Q(:,Z);
l      = l(Z);                % Eigenvalues are sorted by decreasing absolute value

[J,sig] = seleigs(l,alf,prec);% J is the index of the selected eigenvalues
L = l(J);                     % Eigenvalues
Q = Q(:,J);                   % Eigenvectors

W = mapping(mfilename,'trained',{R,Q,L,sig,prec},[],m,sum(sig));
W = setname(W,'Spherical  embedding');
return