Perhaps she's had a lot of anti-semites to play with before.....
Clinical diagnosis is a mixture of art and science. There are a number of ways to look at it, even to the point of dispassionately forcing it into a numerical Bayesian construct with decision analysis. But Linda's example is a great analogy. This is why a clinician knows when she sees someone with chest pain that they are experiencing esophageal spasm and not angina. Of course diagnosis is not always accurate or easy, particularly when there are a lot of co-morbid medical conditions, but
a skillful clinician will outdo a computer algorithm 9 times out of 10, or maybe even 99 times out of a hundred. Think Deep Blue. Similar principles apply.
My own analogy regarding diagnosis is asking someone to say whether the animal they saw was a dog. We all can tell the difference between a dog and the other candidates such as cats, racoons, badgers, rats or whatever. But when you are forced to run through the "checklist" to see if you are correct, you see that other animals might be easily mistaken for the dog - it may have sharp canine teeth, 4 legs, a tail, a hairy coat etc etc. Yet we also know that animals as different as a chiuhahua and a great dane are instantly recognisable as a type of dog using our mental map of dogginess.