New features in IBM SPSS Modeler 18.1
IBM® SPSS® Modeler adds the following features in this release.
TWC source node. The new TWC Input source node imports weather data from The Weather Company, an IBM Business. You can use it to obtain historical or forecast weather data for a location. This can help you develop weather-driven business solutions for better decision-making using the most accurate and precise weather data available. For details, see TWC source node.
Python nodes. The new Python tab on the Nodes palette provides nodes for using Python native algorithms. The SMOTE node, XGBoost Tree node, XGBoost Linear node, and One-Class SVM node are available. These new nodes are supported on Windows 64, Linux64, and Mac. For details, see Python nodes.
Python distribution. If you plan to execute Python/Spark scripts against SPSS Modeler Server (or the local server included with SPSS Modeler Client, which requires Windows 64, Linux64, or Mac), you no longer need to install Python and edit options.cfg as in previous releases. SPSS Modeler now includes a Python distribution. However, if you require a certain module that is not included with the default SPSS Modeler Python distribution, you can go to <Modeler_installation_directory>/python and install additional packages. For more information, see Scripting with Python for Spark.
Language node. The new Language node for IBM SPSS Modeler Text Analytics is a process node that scans source text to determine which human language it is written in and then marks that up in a new field. Primarily designed to be used with large amounts of data, this node is particularly useful when you have more than one language in your data sources and want to process just one language. For details, see Language node.
New build option for Random Trees node. For the Random Trees node, a new advanced build option is available called Number of interesting rules to report. Previously, a maximum of 5 rules were reported. This new option allows you to specify the number of rules to report (minimum of 1, maximum of 1000, with a default of 50). For details, see Random Trees node - advanced.
Import private IBM Cognos TM1 Server views. With the TM1 Import node, you can now import a private view instead of only public views.
Import and export data using R. Using the Custom Dialog Builder for Extensions, you can now create custom nodes and write R scripts to read data from wherever your data source is, and write data out to any data format supported by R. For more information, see Importing and exporting data using R.
Import and export data using Python for Spark. Using the Custom Dialog Builder for Extensions, you can now create custom nodes and write Python for Spark scripts to read data from wherever your data source is, and write data out to any data format supported by Apache Spark. For more information, see Importing and exporting data using Python for Spark.
Analytic Server mode. Under , you can use the Mode tab to switch between traditional mode or Analytic Server mode to filter the nodes and expressions that appear in the user interface. For details, see Setting the mode.
Support for multiple Analytic Servers. If you have multiple Analytic Servers available, you can use the Analytic Server Connection dialog to define more than one server for use in IBM SPSS Modeler. Your administrator may have already set up a default Analytic Server in the <Modeler_install_path>/config/options.cfg file. But you can also use other available servers after defining them. For example, when using the Analytic Server Source and Export nodes, you may want to use different Analytic Server connections in different branches of a stream so that when each branch runs it uses its own Analytic Server and no data will be pulled to the IBM SPSS Modeler Server. Note that if a branch contains more than one Analytic Server connection, the data will be pulled from the Analytic Servers to the IBM SPSS Modeler Server. For more information, including restrictions, see Analytic Server stream properties.
New modes for the Transpose node. On the Settings tab of the Transpose node, a new drop-down has been added to choose a transpose method: Both fields and records, Records to fields, or Fields to records. For details, see Setting options for the Transpose node.
Stream auto save. You can now specify how often to automatically save streams. Go to . For details, see System Options.
Extension nodes. To complement IBM SPSS Modeler and its data mining abilities, the new Extension nodes enable expert users to input their own R scripts or Python for Spark scripts to carry out data processing, modeling building, and model scoring. These Extension nodes replace the R nodes available previously. For more information, see Extension nodes.
Palette management. With Palette Manager (), you can now edit the default node selections for any tab – not just the Favorites tab as in previous releases. For details, see Customizing the Palette Manager.
Improved Microsoft Office support. You can now copy charts as Microsoft Graphic Objects, which allows you to manipulate charts in Microsoft Office. Charts that support this format can be copied to Microsoft Office applications and edited in those applications as native Microsoft Office charts. Because of differences between IBM SPSS Modeler charts and Microsoft Office charts, some features of IBM SPSS Modeler charts are not retained in the copied version. Copying multiple selected charts in Microsoft Office Graphic Object format is not supported. For details, see Printing, saving, copying, and exporting graphs.
CPLEX optimization. The new CPLEX Optimization node provides the ability to use complex mathematical (CPLEX) based optimization via an Optimization Programming Language (OPL) model file. This functionality is also available in the IBM Analytical Decision Management product, but now you can also use the CPLEX node in SPSS Modeler without requiring IBM Analytical Decision Management. For more information, see CPLEX Optimization node.
Online help. In prior releases, the online help system was installed alongside IBM SPSS Modeler. The online help system is now contained within the IBM Knowledge Center. F1 help calls within the IBM SPSS Modeler application now result in the associated help topics being opened in a separate web browser instance that links to the latest version of the documentation in the IBM Knowledge Center.