8
Chapter 2
Defaultse ttings for database connections. You can now specify default settings for SQ L Server
and Oracle database connections, as well as those already supported for IBM DB2 InfoSphere
Warehouse.
Streamproperties andoptimization redesign. The Options tab on the Stream Prop erties dialog box
has been redesigned to group the options into categories. The Optimization optio ns have also
moved from User Options to Stream Properties. For more information, see the topic Setting
Options for Streams in Chapter 5 on p. 54.
Node execution timing. You can now set an option to display indiv idual execution times for
the nodes in a stream. For more info rmation, see the topic Viewing Node Execution Times in
Chapter 5 on p. 67.
Youcan also set an option (time_ecode_execu tion_log)in the server configuration file to record
these execution times in the message log.
Streamparameters in SQLqueries from D atabase source node. You can now include SPSS Modeler
stream parameters in SQL queries that you enter in the Database source node.
ExpressionBu ilder supports in-database functions. If a stream connects to a database through a
Database source node and you use the Expression Builder with a downstream node, you can
include in-database functions from the connected database directly in the expression you are
building. For more information, see the topic Selecting Functions in Chapter 7 on p. 120.
IBMCognos BI nodeenhanceme nts. The Cognos BI source node now supports importing Cognos
list reports as well as data, and additionally supports the use of parameters and filters.
For the Cognos BI source and export nodes, SPSS Modeler now automatically detect s the version
of IBM Cognos BI in use.
Enhancements to Aggregate node. T he Aggregate node now supports several new aggregation
modes for aggregate fields: median, count, variance, and first and thi rd quartiles.
Mergenode supports c onditional merge. You can now perform input record merges that depend on
satisfying a condition. You can specify the condition directly in the node, or build the condition
using the Expression Builder.
Enhancementst o in-database mining nodes for IBM DB2 InfoSphere Warehouse. For in-database
mining with IBM DB2 InfoSphere Warehouse, the ISW Clustering node now supports the
Enhanced BIRCH algorithm in addition to demographic and Kohonen clustering. In addition, the
ISW Association node provides a choice of layout for non-transactional (tabular) data.
Tablecompression for database export. When exporting to a database, you can now specify table
compression options for SQL Server and Oracle database connections, as well as those already
supported for IBM DB2 InfoSphere Warehouse.
Bulkloading for database export. Additional help information is available for database bulk loading
using an external loader program.