IBM Adds Data Mining To DB2 OLAP Server

The combination relies on Hyperion's Essbase OLAP, not IBM's Intelligent Miner.
IBM is adding data-mining capabilities to its DB2 OLAP Server multidimensional analysis software. The move, IBM says, will assist business analysts in such tasks as fraud detection and customer retention by uncovering previously hidden trends and patterns in large volumes of data.

DB2 OLAP Server is a combination of IBM's DB2 relational database and Hyperion Solutions Corp.'s Essbase online analytical processing product. It's used to create data sets that can be analyzed in multiple dimensions, such as determining sales of a particular product in a particular city during a specific month.

The addition of the data-mining software--specifically, deviation-detection algorithms--will let users analyze OLAP data to detect hidden trends, says Jeff Jones, marketing director for IBM data management solutions. ING Antai, a leading insurance company in Taiwan, is using the newly combined technologies for fraud-detection applications. Using the software, the company can analyze fraud claims data in two or three days, a task that previously took several weeks.

The data-mining technology will be built into copies of DB2 OLAP Server beginning Dec. 7. Current owners of DB2 OLAP Server will receive the data-mining module through an upcoming maintenance pack. Jones says IBM's Intelligent Miner data-mining software will still be sold as a separate product.