Improved multi-level association rule in mining algorithm based on a multidimensional data cube
Yingjie Wang, Lili Yu, Qinrun Wen, Congli Liu · 2013
OLAP is a widely used data warehouse practical technology. It not only can be appropriate integrated detailed data, multidimensional view complete data and interactive queries, but also can be used in multi-dimensional analysis, providing analytical modeling tools to generate aggregated data, multidimensional data storage engine for trend analysis and statistical analysis. However, due to too many problems with relying on user input hypothesis, user preconceived problems and limitations severely limit the range of assumptions, which will affect the final conclusion. Compared to OLAP, DM Data mining is based on the various data sources, its analytical process is automatic, users do not need to make clear the problem requires when simply use the tool to dig out from the hidden data collection or potential data model to make predictive analysis. So it is more conducive to find out the unknown but potentially useful information. OLAP and data mining technology are both strengths, but also have shortcomings. If they can combine the advantages of organic development based on OLAP data cubes and data warehouse technology, a new data mining technology will better suit the actual needs. In order to achieve enhanced when combined with OLAP efficiency and flexibility purposes, this paper combines technology and association rule in mining algorithm together, and conduct an appropriate improvements cube at the same time.