Peter J. Haas
Stochastic Models
A sampling-based algorithm for discovering association rules is presented. The finding associations from sampled transactions (FAST) algorithm is a refined sampling-based mining algorithm that is distinguished from other algorithms by its novel two-phase approach to sample collection. Results showed that the FAST is a promising algorithms in situations where good approximate answers suffice and the interactive, real time response is essential.
Peter J. Haas
Stochastic Models
Graham Cormode, Minos Garofalakis, et al.
Foundations and Trends in Databases
Eirinaios Michelakis, Rajasekar Krishnamurthy, et al.
SIGMOD/PODS 2009
Ahmed Elgohary, Matthias Boehm, et al.
CACM