Talk:Learning and analyzing Bayesian networks with Genie


 * in the example with Polly Popular, it seems that the graph described is acceptance->GPA, implying that if Polly accepts, then your GPA will change. Is this the direction of causation that you are going for?  This would imply that Polly is a very popular study partner and tutor.  From the verbal description though it sounds like the direction of causation is reversed (Polly favors invitations from people with high GPAs).
 * same with example 1: temperature of feed influences pressure influences the product, but the graph is backwards.
 * Note that for both of these cases, the inference is correct even if the direction is wrong as these are equivalent networks (i.e. use Bayes rule and you can show they are identical in both directions)


 * Sages corner: getting a job a BP: What happens to the people who are neither accepted or rejected? i.e. if you have a shell internship you will be accepted 20% of the time and rejected 40% of the time, but what happens the other 40% of the time?


 * sages corner: Bayesian networks with Genie: here too the example topology is reversed. OSU's football status is not determined by Michigan's coach, but the reverse may be true.