Mining generalized association rules with fuzzy taxonomicstructures

摘要

Data mining is a key step of knowledge discovery in databases.Usually, Srikant and Agrawal's (1995) algorithm is used for mininggeneralized association rules at all levels of presumed exact taxonomicstructures. However, in many real-world applications, the taxonomicstructures may not be crisp but fuzzy. This paper focuses on the issueof mining generalized association rules with fuzzy taxonomic structures.Particular attention is paid to extending the notions of the degree ofsupport, the degree of confidence and the R-interest measure. Thecomputation of these degrees takes into account the fact that thereexists a partial belonging of any two item sets in the taxonomyconcerned. Finally, a simplified example is given to help illustrate theideas