wolffd@0: function bnet = mk_asia_bnet(CPD_type, p, arity)
wolffd@0: % MK_ASIA_BNET Make the 'Asia' bayes net.
wolffd@0: %
wolffd@0: % BNET = MK_ASIA_BNET uses the parameters specified on p21 of Cowell et al, 
wolffd@0: % "Probabilistic networks and expert systems", Springer Verlag 1999.
wolffd@0: % 
wolffd@0: % BNET = MK_ASIA_BNET('cpt', p) uses random parameters drawn from a Dirichlet(p,p,...)
wolffd@0: % distribution. If p << 1, this is nearly deterministic; if p >> 1, this is nearly uniform.
wolffd@0: % 
wolffd@0: % BNET = MK_ASIA_BNET('bool') makes each CPT a random boolean function.
wolffd@0: %
wolffd@0: % BNET = MK_ASIA_BNET('gauss') makes each CPT a random linear Gaussian distribution.
wolffd@0: %
wolffd@0: % BNET = MK_ASIA_BNET('orig') is the same as MK_ASIA_BNET.
wolffd@0: %
wolffd@0: % BNET = MK_ASIA_BNET('cpt', p, arity) can specify non-binary nodes.
wolffd@0: 
wolffd@0: 
wolffd@0: if nargin == 0, CPD_type = 'orig'; end
wolffd@0: if nargin < 3, arity = 2; end
wolffd@0: 
wolffd@0: Smoking = 1;
wolffd@0: Bronchitis = 2;
wolffd@0: LungCancer = 3;
wolffd@0: VisitToAsia = 4;
wolffd@0: TB = 5;
wolffd@0: TBorCancer = 6;
wolffd@0: Dys = 7;
wolffd@0: Xray = 8;
wolffd@0: 
wolffd@0: n = 8;
wolffd@0: dag = zeros(n);
wolffd@0: dag(Smoking, [Bronchitis LungCancer]) = 1;
wolffd@0: dag(Bronchitis, Dys) = 1;
wolffd@0: dag(LungCancer, TBorCancer) = 1;
wolffd@0: dag(VisitToAsia, TB) = 1;
wolffd@0: dag(TB, TBorCancer) = 1;
wolffd@0: dag(TBorCancer, [Dys Xray]) = 1;
wolffd@0: 
wolffd@0: ns = arity*ones(1,n);
wolffd@0: if strcmp(CPD_type, 'gauss')
wolffd@0:   dnodes = [];
wolffd@0: else
wolffd@0:   dnodes = 1:n;
wolffd@0: end
wolffd@0: bnet = mk_bnet(dag, ns, 'discrete', dnodes);
wolffd@0: 
wolffd@0: switch CPD_type
wolffd@0:   case 'orig', 
wolffd@0:     % true is 2, false is 1
wolffd@0:     bnet.CPD{VisitToAsia} = tabular_CPD(bnet, VisitToAsia, [0.99   0.01]);
wolffd@0:     bnet.CPD{Bronchitis} = tabular_CPD(bnet, Bronchitis, [0.7 0.4   0.3 0.6]);
wolffd@0:     % minka: bug fix
wolffd@0:     bnet.CPD{Dys} = tabular_CPD(bnet, Dys, [0.9 0.2 0.3 0.1   0.1 0.8 0.7 0.9]);
wolffd@0:     bnet.CPD{TBorCancer} = tabular_CPD(bnet, TBorCancer, [1 0 0 0   0 1 1 1]);
wolffd@0:     % minka: bug fix
wolffd@0:     bnet.CPD{LungCancer} = tabular_CPD(bnet, LungCancer, [0.99 0.9  0.01 0.1]);
wolffd@0:     bnet.CPD{Smoking} = tabular_CPD(bnet, Smoking, [0.5 0.5]);
wolffd@0:     bnet.CPD{TB} = tabular_CPD(bnet, TB, [0.99 0.95  0.01 0.05]);
wolffd@0:     bnet.CPD{Xray} = tabular_CPD(bnet, Xray, [0.95 0.02  0.05 0.98]);
wolffd@0:  case 'bool',
wolffd@0:   for i=1:n
wolffd@0:     bnet.CPD{i} = boolean_CPD(bnet, i, 'rnd');
wolffd@0:   end
wolffd@0:  case 'gauss',
wolffd@0:   for i=1:n
wolffd@0:     bnet.CPD{i} = gaussian_CPD(bnet, i, 'cov', 1*eye(ns(i)));
wolffd@0:   end
wolffd@0:  case 'cpt',
wolffd@0:   for i=1:n
wolffd@0:     bnet.CPD{i} = tabular_CPD(bnet, i, p);
wolffd@0:   end
wolffd@0: end
wolffd@0: 
wolffd@0:   
wolffd@0: