Systematic Approach for Optimizing Complex Mining Tasks on Multiple Databases

It has been well recognized that data mining is an interactive and iterative process. In order to support this process, one of the long-term goals of data mining research has been to build a Knowledge Discovery and Data Mining System (KDDMS). Along this line, much research has been done to provide database support for mining operations. However, the focus in these efforts has typically been on mining a single dataset. In many situations, such as in a data warehouse, the user usually has a view of multiple datasets collected from different data sources. In such scenarios, comparing the patterns from different datasets and understanding their relationships can be an extremely important part of the KDD process. This requires support for complex queries on multiple datasets in a KDDMS. This paper addresses the new functionality and optimizations required for the above process, specifically focusing on frequent itemset mining. We make the following contributions: 1) We present an SQL-base...

• Type: Conference
• Year: 2006
• Where: ICDE
• Authors: Ruoming Jin, Gagan Agrawal
• Keywords: Data Mining, Data Mining System, Database, ICDE 2006, Multiple Datasets