Discussions would be more focused if participants were aware of the AFC test -- Abstraction-Filtration-Comparison

https://en.wikipedia.org/wiki/Abstraction-Filtration-Compari...

Will that test be applied here?

Does it need to be revised in light of deep learning?