The use of Bayesian Statistics to "train" spam filters has been a productive one. Spammers can disguise a lot of things, but they can't really disguise their message. Paul Graham, who apparently made the idea more popular, discusses this concept at length.
I think I read somewhere that someone was patenting the idea, or one similarly related. While I'm tempted to digress about the problems with the Patent Office these days, I won't, since plenty of others are doing this.
Anyways, here's other ideas about how to use Bayesian Statistics in other areas. I figured that I don't want to spend the tens of thousands of dollars submitting a patent application, so at least I can "open source" my ideas, and provide for the common good...
Can be used to build a system to facilitate database cleanup, by identifing good or bad database entries.
Can be used to build a system to identify more profitable prospective customers, by identifying good or bad customers.
Can be used to build a system to do "profiling" of suspected criminals or terrorists, by identifying characteristics of bad guys and good guys.
Any others? Post your ideas in the comments box below!
Comments