OLAP Miner User's Guide


Getting started with OLAP Miner

IBM DB2 OLAP Server Miner enables you to examine your OLAP cubes (multidimensional databases) more thoroughly and easily by finding atypical, or deviant, values automatically. Instead of searching through your data manually to find data values that might be higher or lower than you expected, OLAP Miner can do this work for you. The OLAP Miner algorithm is optimized so that it can be deployed on large OLAP cubes.

You can search specific areas (subcubes) of your cubes by creating deviation detection definitions, which specify the way in which OLAP Miner examines the data that you selected for the subcube. OLAP Miner provides an easy-to-use wizard that helps you create deviation detection definitions.

After you create a deviation detection definition, OLAP Miner applies a deviation detection algorithm to the data to find deviations. After the mining run is finished, you can view a simple list of deviations or view the deviations in the Deviation Viewer. From the Deviation Viewer, you can see the deviations in the context of other surrounding values. For more information about how the OLAP Miner algorithm finds deviations, see Appendix A, How OLAP Miner works.

You can also view deviations in a standard spreadsheet application. If you have the DB2 OLAP Server spreadsheet add-in, you can use all of the features of that product to view or report deviations. See the OLAP Miner online help for information about creating deviation detection definitions and viewing mining results.


[ Top of Page | Previous Page | Next Page | Table of Contents | Index ]