Watson Studio, Watson ML, Watson OpenScale

 View Only

General Availability of SPSS Modeler 18.2, C&DS 8.2 and Analytic Server 3.1.2

By Stevan Slusher posted Tue December 04, 2018 03:56 PM

Today we released three new IBM SPSS Data Science Offerings: IBM SPSS Modeler 18.2, Analytic Server 3.1.2 and Collaboration & Deployment Services 8.2. Throughout the last 12 months we have taken feedback in multiple parts of the internet universe (G2Crowd, PAT, Gartner, IT Central Station, Trust Radius, Medallia, etc.). This release, we took many of your requests very seriously. We saw the two highest product drawbacks were 1. Dated user interface, and 2. Lacking modern data visualization capabilities.

This release incorporates three themes:
1. Modernizing the User Interface and Visualizations
2. Bringing Analytics to the Data
3. Embracing & Extending Open Source

Modernizing the User Interface & Visualizations
This release includes an updated look and feel of the user interface as well as added new visualization capabilities. There are four color schemes to choose from (Standard, Classic, Windows & Analytics Carbon). Additionally, you are now able to show stream and SuperNode Mark-ups, which includes comments, model links, and highlighting scoring branches.

Even more, now you are now able to examine a stream’s data. By right-clicking on a node and clicking on 'view data' you can now build twenty-four different charts such as scatter plots, line charts, histograms, treemaps, etc.. Note that Graphboard is still within SPSS Modeler 18.2.

Bringing Analytics to the Data
IBM SPSS Modeler 18.2 extends its in-database mining capabilities. The same algorithms that were available for Netezza are now available for Db2 Warehouse. Using a GUI, Modeler customers can now build and deploy models using the Decision Trees, K-means, Naive Bayes, Bayes Net, KNN, Divisive Clustering, PCA, Regression Tree, Linear Regression, Time Series, Generalized Linear, Two-step algorithms.

Embracing & Extending Open Source
IBM SPSS Modeler 18.2 is shipping with four new Python nodes:
  • Guassian Mixture: For more information see here and here

  • Kernel Density Estimation (KDE) Modeling: For more information see here

  • Kernel Density Estimation (KDE) Simulation: For more information see here

  • Hierarchical Density Based Spatial Clustering (HDBSCAN): For more information see here

    Additional details:

    JSON Import/Export
    You can now import and export JSON data sources. You can access these with two new nodes, JSON Import & JSON Export, and specify the JSON data source to import and where to export.
  • Use the JSON source node to import data from a JSON file into an SPSS® Modeler stream, using UTF-8 encoding. Data in a JSON file can be in form of object, array, or value. This JSON source node only supports reading in an array of objects, and the object can't be nested.

  • Use the JSON export node to output data in JSON format, using UTF-8 encoding. Optionally, you can also create a JSON source node to read the exported data back into the stream.

    SPSS Modeler Text Analytics Enhancements:
  • Import SPSS Text Analytics for Surveys projects (.tas)

  • Flags are now available in the Data pane. You can flag documents with a "complete" flag or an "important" flag. See Flagging responses.

  • Extracted concept results have been improved

  • Empty records are handled the same was as they are in SPSS Text Analytics for Surveys . For example, with an Excel source file, empty records are now kept as part of the text.

  • New Force-In and Force-Out options are available in the Data pane to force records into or out of a category. See Forcing Responses into Categories.

  • Type Reassignment Rules (TRRs) are now available. TRRs transform a sequence of types, macros, and/or tokens into a new concept with a specific type. They can be used in Opinions templates to catch opinions with a change in polarity. For details, see Type Reassignment Rules.

    If you want to learn more about how IBM is solving the toughest challenges in data science and machine learning, come see us at Think 2019! Join the world's top business leaders, visionaries, and innovators at Think 2019 in San Francisco. Register now!

    Also, Join me and the rest of the SPSS team for the featured SPSS Modeler sessions:
  • What's New with IBM SPSS Modeler: Thursday, February 14th 9:30 AM - 10:10 AM | Session ID: 6963A

  • Operationalize Data Science with IBM SPSS Modeler: Thursday, February 14th 4:30 PM - 5:10 PM | Session ID: 6966A

  • #AnalyticServer