| Knowledge discovery in database (KDD), or data mining, aims at the discovery of useful information from large database. Finding association rules is an important mining task, aiming at uncovering all frequent patterns among transactions composed of data attributes. Since association rules were introduced, there has been many endeavor on designing efficient algorithms for finding associate rules. We survey the most important algorithms and approaches developed until now. Further more, we develop bounds on the size of a matrix used to represent a given hypergraph as its family of frequent sets. These new bounds can be used to improve algorithms for generating maximal frequent sets. We also give applications of these bounds for the theory of Boolean functions. |
