bow_kernel.m

《模式分析的核方法》一书中的源代码

function K=bow_kernel(S)

%function K=bow_kernel(S)
%
% This function computes the bag of words kernel matrix for the strings in
% S. The case of the letters is ignored.
%
%INPUTS
% S = a cell array containing the strings
%
%OUTPUT
% kernel = the bow kernel
%
%
%For more info, see www.kernel-methods.net

%
%Author: Tijl De Bie, 25/02/2004, adapted (for speed-up) on 21/10/04.

nS=length(S);
K=zeros(nS,nS);

disp('Build dictionary')
dict={};
for k=1:nS
    s=S{k};
    Swords_long=sort(lower(strread(s,'%s')));
    [Swords{k},indi,indj]=unique(Swords_long);
    ns{k}=diff([0;indi]);
    dict=union(dict,Swords{k});
    lengths(k)=length(Swords{k});
end

disp('Compute kernel')
ndict=length(dict);
endind=ones(nS,1);
for k=1:ndict
    inds=zeros(nS,1);
    d=dict{k};
    inds=[];
    vec=[];
    for i=1:nS
        if endind(i)<=lengths(i)
            t=strcmp(Swords{i}{endind(i)},d);
            if t
                inds=[inds;i];
                vec=[vec;ns{i}(endind(i))];
                endind(i)=endind(i)+1;
            end
        end
    end
    K(inds,inds)=K(inds,inds)+vec*vec';
end