This two-day course is a follow-up course to the Introduction to IBM SPSS Modeler and Data Mining course, and is designed for anyone who wishes to become familiar with the full range of techniques available in IBM SPSS Modeler for data mining.

Course Outline

This class will show you how to use IBM SPSS Modeler to automate the building of predictive models. The course will show you how to build predictive models for customer behavior and build customer segmentation using various cluster models. You will learn how to read data from various sources and automatically prepare data for modeling using a variety of methods. Scoring new data using the model will also be discussed.

Pre-requisites:

  • General computer literacy
  • Some experience using IBM SPSS Modeler including familiarity with the IBM SPSS environment, creating streams, reading data files, and doing simple data exploration and manipulation
  • Prior completion of Introduction to IBM SPSS Modeler and Data Mining is strongly encouraged

High-level Curriculum

Lesson 1: Introduction to Data Mining

  • Automated Data Mining
  • A strategy for data mining: CRISP-DM

Lesson 2: The Basics of Using IBM SPSS Modeler

  • The Modeler User Interface
  • Visual Programming
  • Building Streams with Modeler
  • Modeler Help

Lesson 3: Reading Data Files

  • Reading Data from Statistics Files
  • Defining Field Type
  • Field Role
  • Saving a Modeler Stream

Lesson 4: Data Exploration

  • Missing Data in Modeler
  • The Data Audit Node
  • The Quality Tab
  • Viewing Data with the Table Nodem

Lesson 5: Automated Data Preparation

  • The Type Node
  • Auto Data Prep Node
  • Operation: Using the Auto Data Prep Node

Lesson 6: Data Partitioning

  • Data to Train and Test Models
  • The Partition Node

Lesson 7: Predictor Selection for Modeling

  • The Feature Selection Node
  • Feature Selection Model

Lesson 8: Automated Models for Categorical Targets

  • The Auto Classifier Node
  • Auto Classifier Model

Lesson 9: Model Evaluation

  • Model Predictions with the Analysis Node
  • Selecting the Testing Partition Records
  • Using the Matrix Node for Model Predictions
  • Model Predictions for Categorical Input Fields
  • Model Predictions for Continuous Input Fields
  • Improving the Model

Lesson 10: Automated Models for Continuous Targets

  • Data Preparation Stream to Predict Tenure
  • The Auto Numeric Node
  • Auto Numeric Model
  • Model Predictions with the Analysis Node
  • Selecting the Testing Partition records
  • Model Predictions for Categorical Fields
  • Model Predictions for Continuous Fields

Lesson 11: Deploying Models

  • The Deployment Phase
  • Deploying a Model
  • Exporting Model Results
  • Other Deployment Options

Lesson 12: Course Summary

  • Course Objectives Review/globalli]