and 1996/1997. So why, almost 10 years after the expert
BN was developed, have we returned to this particular
problem? It is because we had a unique opportunity for a
direct comparison between the expert BN and a range of
alternative ML models. Such studies are relatively rare
and the results and lessons learnt should be of interest to
researchers outside of this particular domain (even those
readers who have no interest in Spurs or football in general).
The performance of the expert BN model is compared
with four alternative machine learning (ML)
models: