Download All

Code covered by the BSD License  

Highlights from
NPC algorithm for learning DAG in Bayesian network

image thumbnail
from NPC algorithm for learning DAG in Bayesian network by Guangdi Li
NPC algorithm is designed for learning Bayesian network formed as DAG in 2001, by Steck

CITest_ChiTwoVar( MI, R, M, a ) 
function CI = CITest_ChiTwoVar( MI, R, M, a )
% Input:
% MI is the mutual information of I(X1,X2) or I(X1,X2|others );
% M is the number of sample used
% a is  0.05

%Output:
% if CI = 1,X1 and X2 independent, otherwise,dependent
% Null hypothesis is X1 and X2 are independent

   CI = 0; 
   Threshold = chi2inv( 1-a,R );
   if  Threshold < 2*M*MI      %|| ThresholdLarge < 2*N*MI
       CI = 1;    
   end
end

Contact us at files@mathworks.com