Advanced Predictive Modeling Using IBM SPSS Modeler (v18.1.1)

Audience

• Business Analysts
• Data Scientists
• Users of IBM SPSS Modeler responsible for building predictive models

Prerequisites

• Familiarity with the IBM SPSS Modeler environment (creating, editing, opening, and saving streams).
• Familiarity with basic modeling techniques, either through completion of the courses Predictive Modeling for Categorical Targets Using IBM SPSS Modeler and/or Predictive Modeling for Continuous Targets Using IBM SPSS Modeler, or by experience with predictive models in IBM SPSS Modeler.

Course topics

1. Preparing data for modeling

• Address general data quality issues
• Handle anomalies
• Select important predictors
• Partition the data to better evaluate models
• Balance the data to build better models

2. Reducing data with PCA/Factor

• Explain the idea behind PCA/Factor
• Determine the number of components/factors
• Explain the principle of rotating a solution

3. Creating rulesets for flag targets with Decision List

• Explain how Decision List builds a ruleset
• Use Decision List interactively
• Create rulesets directly with Decision List

4. Exploring advanced supervised models

• Explain the principles of Support Vector Machine (SVM)
• Explain the principles of Random Trees
• Explain the principles of XGBoost

5. Combining models

• Use the Ensemble node to combine model predictions
• Improve model performance by meta-level modeling

6. Finding the best supervised model

• Use the Auto Classifier node to find the best model for categorical targets
• Use the Auto Numeric node to find the best model for continuous targets