The HPBNET Procedure

This example shows how you can use PROC HPBNET to learn a naive Bayesian network for the Iris data available in the Sashelp library. The following statements specify MAXPARENTS=1, PRESCREENING=0, and VARSELECT=0 to request that PROC HPBNET use only one parent for each node and use all the input variables.

 proc hpbnet data=sashelp.Iris numbin=3 structure=Naive maxparents=1
         prescreening=0 varselect=0;
     target Species;
     input PetalWidth PetalLength SepalLength SepalWidth/level=INT;
     output network=network;
 run;

The following statements produce Figure 4, which shows the network structure. In the structure, Species is the parent of PetalWidth, PetalLength, SepalLength, and SepalWidth.

 proc print data=network noobs label;
     var _parentnode_ _childnode_;
     where _type_="STRUCTURE";
 run;