Publication
Proceedings - International Conference on Data Engineering
Paper
FAST: A new sampling-based algorithm for discovering association rules
Abstract
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.