How would you go about building a Bayesian gaydar?
Put a human with good social skills in a box, expose it to a representative sample of people of various sexualities and reward it when it guesses right; the human brains social functionality is a very powerful specialized Bayesian engine. :p
Alternatively, just take your own brain and expose it to a large representative sample of people of varying sexualities and only check what they were afterwards. Not quite as technically powerful, but more portable and you get some extra metadata.
This is a thread where people can ask questions that they would ordinarily feel embarrassed for not knowing the answer to. The previous thread is at close to 500 comments.