Mining Generalized Association Rules-英文文献.pdfVIP

Mining Generalized Asso ciation Rules Ramakrishnan Srikant Rakesh Agrawal IBM Almaden Research Center San Jose CA fsrikantragrawal galmadenibm com Intro duction Abstract Data mining also known as knowledge discovery in databases has b een recognized as a new area for We intro duce the problem of mining general database research The area can b e dened as e ized asso ciation rules Given a large database ciently discovering interesting rules from large collec of transactions where each transaction con tions of data sists of a set of items and a taxonomy isa The problem of mining asso ciation rules was intro hierarchy on the items we nd asso ciations duced in Given a set of transactions where each b etween items at any level of the taxonomy transaction is a set of items an asso ciation rule is For example given a taxonomy that says that an expression X Y where X and Y are sets of jac kets isa outerwear isa clothes we may items The intuitive meaning of such a rule is that infer a rule that p eople who buy outerwear transactions in the database which contain the items tend to buy sho es This rule may hold even in X tend to also contain the items in Y An example if rules that p eople who buy jackets tend to of such a rule might b e that of customers who buy sho es and p