Machine Learning and AI Foundations: Clustering and Association

Introduction

• Welcome
• What you should know
• Using the exercise files
• What is unsupervised machine learning?

1. What Is Cluster Analysis?

• Looking at the data with a 2D scatter plot
• Understanding hierarchical cluster analysis
• Running hierarchical cluster analysis
• Interpreting a dendrogram
• Methods for measuring distance
• What is k-nearest neighbors?

2. K-Means

• How does k-means work?
• Which variables should be used with k-means?
• Interpreting a box plot
• Running a k-means cluster analysis
• Interpreting cluster analysis output
• What does silhouette mean?
• Which cases should be used with k-means?
• Finding optimum value for k: k = 3
• Finding optimum value for k: k = 4
• Finding optimum value for k: k = 5
• What the best solution?

3. Visualizing and Reporting Cluster Solutions

• Summarizing cluster means in a table
• Traffic Light feature in Excel
• Line graphs

4. Cluster Methods for Categorical Variables

• Relating clusters to categories statistically
• Relating clusters to categories visually
• Running a multiple correspondence analysis
• Interpreting a perceptual map
• Using cluster analysis and decision trees together
• A BIRCH/two-step example
• A self organizing map example

5. Anomaly Detection

• The k = 1 trick
• Anomaly detection algorithms
• Using SOM for anomaly detection

6. Association Rules and Sequence Detection

• Intro to association rules and sequence analysis
• Running association rules
• Some association rules terminology
• Interpreting association rules
• Putting association rules to use
• Comparing clustering and association rules
• Sequence detection

Conclusion

• Next steps