Partitioned approach to association rule mining over multiple databases

Database mining is the process of extracting interesting and previously unknown patterns and correlations from data stored in Data Base Management Systems (DBMSs). Association rule mining is the process of discovering items, which tend to occur together in transactions. If the data to be mined were stored as relations in multiple databases, instead of moving data from one database to another, a partitioned approach would be appropriate. This paper addresses the partitioned approach to association rule mining for data stored in multiple Relational DBMSs. This paper proposes an approach that is very effective for partitioned databases as compared to the main memory partitioned approach. Our approach uses SQL-based K-way join algorithm and its optimizations. A second alternative that trades accuracy for performance is also presented. Our results indicate that beyond a certain size of data sets, the accuracy is preserved in addition to improving performance. Extensive experiments have been performed and results are presented for the two partitioned approaches using IBM DB2/UDB and Oracle 8i.