Machine Learning and AI Foundations: Decision Trees with KNIME

Expand your data science skills and establish a strong foundation in codeless machine learning.

Introduction

• The basics of decision trees
• What you should know
• How to use the practice files

1. Introducing Decision Trees

• What is a decision tree?
• The pros and cons of decision trees
• Introducing KNIME
• A quick review of machine learning basics with examples
• An overview of decision tree algorithms

2. Introducing the C5.0 Algorithm

• Ross Quinlan, ID3, C4.5, and C5.0
• Understanding the entropy calculation
• How C4.5 handles missing data
• The Give Me Some Credit data set
• Working with the prebuilt example
• KNIME settings for C4.5
• How C4.5 handles nominal variables
• How C4.5 handles continuous variables
• Equal size sampling
• A quick look at the complete C4.5 tree
• Evaluating the accuracy of your C4.5 tree
• When to turn off pruning

3. Introducing Classification Trees

• Introducing Leo Breiman and CART
• What is the Gini coefficient?
• How CART handles missing data using surrogates
• Changing the settings in KNIME
• How CART handles nominal variables
• A quick look at the complete CART tree
• Evaluating the accuracy of your CART tree

4. Introducing Regression Trees

• MPG data set
• The regression tree prebuilt example
• The math behind regression trees
• How RT handles nominal variables
• Ordinal variable handling
• Closer look at a full regression tree
• KNIME's missing data options for regression trees
• Line plot
• Accuracy

Conclusion

• Next steps