Introduction to IBM SPSS Text Analytics for IBM SPSS Modeler


Course Overview

Getting usable analytics results from unstructured, text-based data requires a very different set of strategies than traditional numeric data does. Despite this, however, there are many insights you can gain from text-based sources through the right application of predictive technology.

Ironside’s 2-day Introduction to IBM SPSS Text Analytics for IBM SPSS Modeler introduces you to the text analytics module available in Modeler. You’ll learn the steps involved in working with text-based data from the initial read all the way to final category creation. By the course’s conclusion, you’ll have all the knowledge you need to generate powerful predictive models from text that will lead to actionable intelligence for your organization.


This course is intended for anyone who needs to report on or generate predictive models from text data. Students should have knowledge equivalent to having taken the IBM SPSS Modeler Fundamentals course, and practical experience coding is beneficial but not required.

Course Goals

  • Introduce the concept of text mining and how it differs from standard data mining.
  • Understand how SPSS Modeler reads text and the options you have for manipulating text data within it.
  • Recognize the different text mining model approaches available to you and become comfortable using them.

High-Level Curriculum

  • Learn the processes for text mining, the steps in a text mining project, and its relationship to the standard data mining/CRISP-DM process.
  • Recognize and work with the text mining nodes available in SPSS Modeler and read text from documents and web feeds into Modeler.
  • Describe the concepts behind linguistic analysis and develop a text mining concept model
  • Use the Interactive Workbench to extract types and concepts and update the modeling node.
  • Edit the linguistic resources available to you, including preparation, synonym definitions, exclusion definitions, and text re-extraction.
  • Fine tune your resources with advanced functionality like fuzzy grouping exceptions, adding non-linguistic entities, extracting non-linguistic entities, and forcing words to take particular parts of speech.
  • Perform text link analysis using the appropriate node and the visualization pane.
  • Understand clustering concepts and create clusters and categories from clusters.
  • Become familiar with the different categorization techniques available.
  • Create and assess categories both manually and automatically, use conditional rules to create categories, and create text analysis packages.
  • Manage your linguistic resources using the Template Editor to build and manage libraries, templates, and backup resources.
  • Use text mining models containing quantitative and qualitative data and score new data.