Composition with target constraints
Marcelo Arenas, Ronald Fagin, et al.
ICDT 2010
We introduce backoff processes, an idealized stochastic model of browsing on the World Wide Web, which incorporates both hyperlink traversals and use of the "back button." With some probability the next state is generated by a distribution over out-edges from the current state, as in a traditional Markov chain. With the remaining probability, however, the next state is generated by clicking on the back button and returning to the state from which the current state was entered by a "forward step." Repeated clicks on the back button require access to increasingly distant history. We show that this process has fascinating similarities to and differences from Markov chains. In particular, we prove that, like Markov chains, back-off processes always have a limit distribution, and we give algorithms to compute this distribution. Unlike Markov chains, the limit distribution may depend on the start state.
Marcelo Arenas, Ronald Fagin, et al.
ICDT 2010
Ronald Fagin
Theoretical Computer Science
Yehoshua Sagiv, Claude Delobel, et al.
Journal of the ACM
Ronald Fagin, Phokion G. Kolaitis, et al.
ACM TODS