getarcs {rWMBAT} | R Documentation |
By comparing mutual information between two variables to thresholds determined seperately, this function declares there to be an arc in a Bayesian network. Arcs are stored in an adjacency matrix, described below.
getarcs(mvc, vcthreshold, mvv, vvthreshold)
mvc |
double vector array with mutual information between variables and the class (variables and other variables). The (i,j) entries of mvv are MI(Vi,Vj) |
vcthreshold |
scalar threshold used to test for existence linkz |
mvv |
double vector matrix |
vvthreshold |
scalar threshold used to test for existence linkz |
The primary tests are MI[Vi,Cj]>>vcthreshold (tests for links between Vi and the class) MI[Vi,Vj]>>vvthreshold (tests the links between variables)
logical matrix whose entries "1" at (i,j) mean "an arc exists from the Bayesian network node Vi to Vj." The class variable C is added at row (number of V's + 1). "0" values mean no arc
Karl Kuschner, Qian Si and William Cooke, College of William and Mary, Dept. of Physics, 2009
http://kwkusc.people.wm.edu/dissertation/dissertation.htm
data(mi_vc, vcthreshold, mi_vv, vvthreshold) # load the example input data from the package adjmat<-getarcs(mi_vc, vcthreshold, mi_vv, vvthreshold)