Home > brede > brede_mat_mma.m

brede_mat_mma

PURPOSE ^

brede_mat_mma - Multinomial mixture analysis

SYNOPSIS ^

function [T, A, B ] = brede_mat_mma(M, varargin)

DESCRIPTION ^

 brede_mat_mma        - Multinomial mixture analysis

       [U,S,V] = brede_mat_mma(M)

       Input:  M    Matrix as a 'mat' structure

       Output: U    Matrix in 'mat' structure
               S    Diagonal matrix
               V    Matrix in 'mat' structure

       Estimation of the multinomial mixture model (mixture of
       multinomials). 

       It presently seems not to work well. It is suggested to use
       the non-negative matrix factorization brede_mat_nmf instead.

       Based on update equations suggested by Rigouste, Cappe and
       Yvon in 'Evaluation of a Probabilistic Method for Unsupervised
       Text Clustering'.

       Example: 
         f = fullfile(fileparts(which('brede')), 'xml', 'wobibs.xml');
         B = brede_read_xml(f, 'output', 'collapsesecond');
         M = brede_bib_bib2mat(B, 'type', 'abstract');
         M = brede_mat_elimsingle(M)
         M = brede_mat_elimstop(M, 'filename', 'stop_english1.txt')
         M = brede_mat_elimstop(M, 'filename', 'stop_medline.txt')
         M = brede_mat_elimstop(M, 'filename', 'stop_brainanatomy.txt')
         M = brede_mat_elimstop(M, 'filename', 'stop_meshcommon.txt')
         M = brede_mat_elimstop(M, 'filename', 'stop_pubmed_neg1.txt')
         [U,S,V] = brede_mat_mma(M);
         brede_mat_2mat2html(U, V)
         web(['file:///' which('Mat.html')]);

       See also BREDE, BREDE_MAT, BREDE_MAT_ICA, BREDE_MAT_KMC,
                BREDE_MAT_NMF, BREDE_MAT_SVD.

 $Id: brede_mat_mma.m,v 1.1 2005/09/27 20:24:27 fnielsen Exp $

CROSS-REFERENCE INFORMATION ^

This function calls: This function is called by:
Generated on Fri 27-Nov-2009 18:11:22 by m2html © 2005