AOL released the list of searches created by users, purportedly anonymized. You didn't have access to the index of their crawler. Also, their subset was reall

AOL released the list of searches created by users, purportedly anonymized. You didn't have access to the index of their crawler. Also, their subset was really small. I'm talking about a continuously updated stream. I have started learning about complex event processing (CEP) and am wondering if you could do anything with that as a model tied to suggesting related links/sites.