Bayesian Filter Follow up

I found out what the deal with the Bayesian Filter is. I emailed my Unix professor and asked her if she knew anything about it. She said that she knew that a Bayesian Filter HAD in fact been installed on SRU's mail server which is why I was getting my spam marked. I also asked if there was a place to send false positives so that the filter could be trained and she said that the false positive problem was very common around campus and was the reason why the junk mail was not deleted server side. She said she didn't know if the person in charge was interested in training the filter or not. My guess is not because that would take some time to process everyones false positives. But it's still pretty useful because I can tell outlook to delete all emails with "[SPAM]" in the title, except for those that come from people in my address book. Good stuff. They use the same "trainable" software with bank loans. They process past loans and train the software to identify which applicants will be able to pay off their loan and which ones won't. I think the concept of trainable software is called using a "neural network" or something akin to that. It's pretty interesting.