vivster on 11 July 2018
So this experiment was conducted by someone who has vested interest in proving a genetic sexuality hormone theory. Can't see any possible experimentation bias here.
There is no detection here, only good guesses, most likely with biased samples. A learning algorithm is only as good as the material you feed it with. Try feeding it with pictures of homosexual people that contradict "typical" facial features of homosexuals and see the accuracy plummet. Since there are absolutely no hard physical clues for someone's sexuality this algorithm's gaydar is just as good as anyone else's.
Please stop spreading FUD.
Last edited by vivster - on 11 July 2018If you demand respect or gratitude for your volunteer work, you're doing volunteering wrong.
