nDenominator can be viewed as a normalization constant α
n
nP(Cavity | toothache) = α · P(Cavity, toothache)
n= α · [P(Cavity, toothache, catch) + P(Cavity, toothache, Øcatch)]
n= α · [<0.108,
0.016> + <0.012, 0.064>]
n= α · <0.12, 0.08>
= <0.6, 0.4>
n
nGeneral idea: compute
distribution on query variable by fixing evidence variables and summing over hidden variables