作者: Riadh Ben Messaoud , Omar Boussaid , Sabine Rabaseda
DOI: 10.1109/INNOVATIONS.2006.301947
关键词: Data cube 、 Aggregate (data warehouse) 、 Association rule learning 、 Computer science 、 Online analytical processing 、 Data mining 、 Data structure 、 Data stream mining 、 Context (language use) 、 Lift (data mining)
摘要: On-line analytical processing (OLAP) provides tools to explore data cubes in order extract interesting information. Nevertheless, OLAP is not capable of explaining relationships that could exist within data. Association rules are one kind mining techniques which finds associations among In this paper, we propose a framework for association from according sum-based aggregate measure more general than frequencies provided by the COUNT measure. Our process guided meta-rule context driven analysis objectives and exploits measures revisit definition support confidence. We also evaluate interestingness mined Lift Loevinger criteria an algorithm inter-dimensional directly multidimensional structure