346 processing im functions all im functions can be

This preview shows page 64 - 69 out of 260 pages.

We have textbook solutions for you!
The document you are viewing contains questions related to this textbook.
CMPTR
The document you are viewing contains questions related to this textbook.
Chapter 16 / Exercise 17
CMPTR
Pinard/Romer/Morley
Expert Verified
3.4.6 Processing IM Functions All IM functions can be customized using two levels of expertise. Users who are not experts can accept the defaults and suppress advanced settings. However, experienced users who want to fine-tune their application have the ability to customize all settings according to their requirements. You can additionally define the mode of your statistical and mining functions. Possible modes are:
We have textbook solutions for you!
The document you are viewing contains questions related to this textbook.
CMPTR
The document you are viewing contains questions related to this textbook.
Chapter 16 / Exercise 17
CMPTR
Pinard/Romer/Morley
Expert Verified
The Products and Their Construction 49 • Training mode. In training mode a mining function builds a model based on the selected input data. • Clustering mode. In clustering mode, the clustering functions build a model based on the selected input data. • Test mode. In test mode, a mining function uses new data with known results to verify that the model created in training mode produces consistent results. • Application mode. In application mode, a mining function uses a model created in training mode to predict the specified fields for every record in the new input data. 3.4.7 Creating and Visualizing the Results Information that has been created using statistical or mining functions can be saved for further analysis in the form of result objects. Figure 10 illustrates some of the results generated by the clustering function.
50 Getting Started with Data Warehouse and Business Intelligence Figure 10. Sample Clustering Visualization Result objects can be used in several ways: • To visualize or access the results of a mining or statistical function • To determine what resulting information you want to write to an output data object • As input when running a mining function in test mode to validate the predictive model representation by the result • As input when running a mining function in application mode to apply the model to new data 3.4.8 Creating Data Mining Operations The IM provides a means to construct data mining operations as a sequential series of related data preparation, statistical, and mining functions. IM will execute the objects in the order in which they have been specified. A
The Products and Their Construction 51 sequence can contain other sequences. You can construct standard sequences that can be reused in similar data mining operations forming part of a more complex sequence. In IM, sequence setting objects are created through the sequence task guide. You can select objects from the mining database and place them in the sequence work area setting in the order in which they are to run. In addition you can specify whether the sequence should continue if any of the setting objects fails. Figure 11 illustrates a sequence-settings window. Figure 11. Sequence-Settings Window 3.5 DB2 OLAP Server The IBM DB2 OLAP Server is an online analytical processing (OLAP) server that you can use to create a wide range of multidimensional planning, analysis, and reporting applications. DB2 OLAP Server uses the Essbase OLAP engine developed by Arbor Software Corporation. The OLAP Server uses the Essbase engine for application design and management, data
52

  • Left Quote Icon

    Student Picture

  • Left Quote Icon

    Student Picture

  • Left Quote Icon

    Student Picture