I think the bayesian filter approach (as in email/spam) has one major drawback/shortcoming when used for aggregators:
I'm not only interested in one kind of post/topic, but rather a few categories, that can be varied and very different. And also I might on accident find something totally new, which I'd in the future also like to read... A bayesian filter can't help here.
What is needed is a way of 'more like this'... A way to mark posts I like and have the aggregator show me otheres that are similar...
And a list of categories with a 'sample corpus' for each will not really help... categories are to unflexible and not ad-hoc enough to be really easy to use (need thinking about). Similar entries like I have here are a no-brainer user-wise. Just find a post you like and browse fron there...
read this post
form some very cool ideas...
(using the most intersting tokens to query feedster for other feeds)
[ by Martin>]
similar entries (vs):
similar entries (cg):