Predictive Modeling with IBM SPSS Modeler
IBM SPSS Modeler offers the predictive power of proven statistical algorithms combined with the usability and customizability of an intuitive graphical user interface. SPSS Modeler enables users to proactively add predictive analytics capabilities to their organization. As part of our valuable curriculum of IBM SPSS training, the Ironside Group provides instructor-led classes that introduce users to the advanced predictive modeling capabilities of SPSS Modeler and give them the detailed understanding of these principles necessary to create models that will yield the highest predictive power possible.
This two-day class is suited for users from any industry who have experience using IBM SPSS Modeler and who want to fully leverage the program’s modeling options. During the Predictive Modeling with IBM SPSS Modeler class, students will expand their knowledge of SPSS Modeler by creating sophisticated models from data files using the tools available in the program. At the course’s conclusion, students will have an advanced grasp of the model creation process and will be able to fully capitalize on the modeling tools available in SPSS Modeler. Each topic will be highlighted with in-depth demos, Q&A sessions, and tips and techniques drawn from over 20 years of cross industry experience.
The curriculum is based on the most recent version of IBM SPSS Modeler, and uses the sample data included in the Demo folder. For an even more enriching experience, we can customize course content to any customer’s environment, including but not limited to the tutorials and workshops. All our instructors have over 20 years of real-world experience with SPSS Statistics, advanced statistical concepts, and predictive analytics in general.
- Experience with predictive modeling.
- Completion of Introduction to Data Mining and IBM SPSS Modeler or equivalent experience.
- Intermediate comfort level with the SPSS Modeler program.
- Preparing Data for Modeling
- Searching for Data Anomalies
- Selecting Predictors
- Data Reduction with Principal Components
- Neural Networks
- Support Vector Machines
- Cox Regression
- Time Series Analysis
- Decision Trees
- Linear Regression
- Logistic Regression
- Discriminant Analysis
- Bayesian Networks
- Numeric Predictor Node
- Binary Classifier Node
- Combining Models to Improve Performance
- Getting the Most from Models