SPSS Modeler

Why Upgrade?

With each release, we’ve added a number of helpful features. Take a look at what you’re missing. Then, contact us to upgrade your existing IBM SPSS Modeler software.

IBM SPSS Modeler Version 15.0.0.1 Added:

  • Binning Node: An option has been added to the “Fixed-width binning” mode that causes the same bin intervals to be used for all binned fields
  • Auto-Classifier and Auto-Numeric Model Builder and Model Applier Nodes: An option has been added that generates the individual model outputs, in addition to the output of the model ensemble
  • Evaluation Node: An option has been added that overlays profit criteria for all chart types, where previously this option was only available for the Profit chart type
  • AIX Power performance enhancements (read/write improvements, some algorithms)
  • Improvements to Entity Analytics support within Collaboration and Deployment Services
  • Support for the Windows 8 platform for Modeler clients
  • Additional Data Source Support
    • SAP HANA
    • Greenplum
    • Teradata 14
    • SQL Server 2012 (Windows only)
    • Support for the IBM DB2 ODBC driver

IBM SPSS Modeler Version 15.0 Added:

  • Productivity and User Experience Enhancements including Zoom
  • Enhanced In Database and SQL Pushback Support
  • Scoring Adapters for In Database Analytics
  • Enhanced Aggregation Options
  • Generalized Linear Mixed Models (GLMM)
  • Support for Maps in Graphboard
  • Translation to Brazilian Portuguese
  • Support for Informix Database
  • Entity Analytics (Modeler Premium)
  • Social Network Analysis (Modeler Premium)
  • Text Analytics Performance Enhancements (Modeler Premium)

IBM SPSS Modeler Version 14.2 Added:

  • Support for Netezza Analytics In Database Mining
  • Netezza Data Source Support
  • SQL Pushback Enhancements
  • Support for IBM PowerVM Hypervisor

IBM SPSS Modeler Version 14.1 Added:

  • Cognos Import and Export nodes
  • Database and SQL Pushback Enhancements
  • Updated C5 Algorithm
  • PMML 4.0 Support
  • IBM Classic Federation Server (mainframe data access)
  • SuSE Linux Enterprise Server Support

IBM SPSS Modeler Version 14.0 Added:

  • Boosting and bagging for key algorithms: Neural Net, Linear, C&RT, Chaid and Quest
  • Large dataset processing optimization for key algorithms: Neural Net, Linear, C&RT, Chaid and Quest (Requires IBM SPSS Modeler Server)
  • Interactive ensemble model viewer
  • New Neural Net algorithm that includes multilayer perceptron and radial basis function
  • New method for Linear Regression (Linear) that leverages boosting, bagging and large database processing, along with an automatic data preparation option built-in
  • Consistency between names of variable measurement types and roles definitions, as well as a common look and feel across IBM SPSS Statistics and IBM SPSS Modeler
  • XML Source and Export nodes
  • Additional In Database algorithms (Microsoft Time Series, Microsoft Sequence Clustering and Oracle Attribute Importance)
  • The ability to update database tables in Database Export node
  • Ability to prompt for parameter values at stream run-time
  • Excel Export node enhancements (ability to insert into an existing file)
  • Improved deployment and scoring through introduction of a deployment definition that enables automatic rebuilding of models, specifying scoring branches and ability to define model refresh
  • Support for SSO for Modeler Server (requires IBM SPSS Collaboration and Deployment Services)
  • The ability to launch a model developed in Modeler within Modeler Advantage from Modeler’s interface
  • Support for IBM SPSS Statistics’ custom dialogues to be used from the modeling interface and support for both spv and spw output formats

Version 13.0 (PASW Modeler) Added:

  • Ability to leverage SPSS Statistics software from within the Modeler interface
  • Automated Data Preparation node
  • Data audit node
  • Password-protection for Supernodes
  • New Interactive rule-building algorithm
  • Ability to search for a node within a stream
  • Output Preview
  • Nearest Neighbor algorithm
  • Auto Cluster Node
  • Support for virtualized server environments
  • Enhance productivity and build streams faster because a wider range of scoring and data preparation and data operations are pushed back to the SQL database
  • Visualization enhancements (automatically generate graphs from outputs and publish Modeler output to the Web)

Version 12.0 (Clementine) Added:

  • RFM customer value segmentation scoring technique
  • Cox regression for survival analysis
  • Automated model building for binary ('yes/no') and numeric outcomes with support for frequency, weights, and misclassification costs
  • Ensemble node
  • Support for stratified and clustered sampling
  • Support for propensity scores
  • Support Vector Machine (SVM) node
  • Graphical support for Bayesian network models
  • Interact with graphs via rich data selection tools
  • New Custom tables. Nest, stack, or layer variables in multiple dimensions to display summaries for multiple statistics and display multiple response sets
  • Variable importance charts
  • Extended database optimization
  • Enhanced real-time scoring capabilities

Version 11.0 (Clementine) Added:

  • Parallel execution of entire data mining streams for faster performance, for users of hardware with multiple CPUs
  • Encrypted communication using Secure sockets Layer (SSL)
  • Password protection for nodes, streams, states, or projects
  • Anonymizer node, for creating datasets without any personal data
  • Missing value imputation based on both rule- and algorithm-based methods
  • Interactive filtering of outliers and extreme values
  • Enhanced optimal binning to preserve the strength of the association between two fields
  • SPSS Transform node (requires licensed, installed copy of SPSS) for running SPSS syntax on data within Clementine
  • Enhanced Transform node to support defining and selecting transformations to normalize data across variables in a single step
  • Binary Classifier node supports automatic selection and comparison of models for binary outcomes
  • Self-learning response models, so that models can be re-estimated without re-training on all data
  • Discriminant Analysis node as an alternative or supplement to logistic regression analysis
  • Generalized Linear Models node for greater flexibility in model creation
  • Time-Series Modeling node estimates exponential smoothing and ARIMA models to produce forecasts
  • Decision List algorithm for greater control over model elements
  • Additional algorithms supporting integration with Oracle, IBM, and Microsoft in-database modeling
  • Enhanced Logistic Regression node
  • New graph-creation engine for editing and presentation of result graphs
  • Enhanced SPSS Output node for incorporating SPSS graphs, statistical analysis and reporting within a Clementine process
  • Scenario creation and deployment through SPSS Predictive Enterprise Services (now Collaboration and Deployment Services)
  • Support for PMML 3.1
  • SQL Model Scoring Code export

Contact IBM

Considering a purchase?

  • #

    Putting the Lid on Churn

    Read how XO Communications uses SPSS Modeler to boost customer revenue retention by 60 percent.