A Decomposition Algorithm for Learning Bayesian Networks Based on Scoring Function

Table 2

Scores on Alarm for different dataset size 𝑛: score of original network Sc (𝐺₀), score of network returned by Algorithm 1 Sc (𝐺₁), and score of network returned by K2 algorithm Sc (𝐺₂).

Score	𝑛=500	𝑛=1000	𝑛=2000	𝑛=5000	𝑛=10000
Sc (𝐺0)	−5.2004𝑒+003	−1.0000𝑒+004	−1.9440𝑒+004	−4.7737𝑒+004	−9.4515𝑒+004
Sc (𝐺1)	−4.9831𝑒+003	−1.0652𝑒+004	−2.0502𝑒+004	−4.7558𝑒+004	−9.4514𝑒+004
Sc (𝐺2)	−6.8038𝑒+003	−1.9824𝑒+004	−3.0308𝑒+004	−4.7826𝑒+004	−9.4533𝑒+004