Discovery of multivalueddependenciesfrom database relations isviewed as a search in a hypothesis space de ned according to the generalisation relationship among multivalued dependencies. Two algorithms for the discovery of multivalued dependencies from relations are presented. The topdown algorithm enumerates the hypotheses from the most general to more speci c hypotheses which are checked on the input relation. The bottomup algorithm rst computes the invalid multivalued dependencies. Starting with the most general dependencies, the algorithm iteratively re nes the set of dependencies to conform with each particular invalid dependency. The implementation of the algorithms is analysed and some empirical results are presented.
Iztok Savnik, Peter A. Flach