Introduction to Data Mining
- Explain the stages of the CRISP-DM process model.
- Describe successful data mining projects and the reasons why projects fail.
- Describe the skills needed for data mining.
Working with Streams
- Describe the different areas of the Modeler User Interface.
- Work with nodes and Supernodes.
- Run, open and save a stream.
- Access the help function within Modeler.
Data Mining Tour
- Explain the primary concepts used in data mining.
- Build, evaluate and deploy a model.
- Use the Sort and Filter nodes.
Collecting Initial Data
- Explain the concepts of "data structure", "records", "fields", "unit of analysis", "storage".
- Read data from and export data to various file formats
- Examine the distributions of categorical and continuous fields.
- Explain the most common ways of handling missing data.
- Explain the most common ways of handling outliers.
- Explain how to set Modeler to check data quality and select valid records.
Setting the Unit of Analysis
- Remove duplicate records.
- Aggregate data.
- Create flag fields.
- Restructure continuous fields.
- Add records from multiple datasets into one dataset.
- Add fields from multiple datasets into one dataset.
- Use sampling for testing purposes.
Deriving and Filling Fields
- Use CLEM to transform data.
- Use the Derive node to create a new field, and the Filler node to replace values in a field.
- Explain how to automatically generate a Derive node.
- Use the Reorder node to reorder fields.
Reclassifying and Binning Fields
- Use the Reclassify node.
- Explain how to automatically generate a Reclassify node.
- Use the Binning node.
Looking for Relationships
- Examine the relationship between two categorical fields.
Introduction to Classification
- Differentiate between predictive modeling and other types of modeling.
- Differentiate between various types of predictive models.
- Run CHAID in interactive mode.
- Run CHAID and various other models in automatic mode.
Tillbaka till toppen