4 Association rules

This chapter covers

Association rules
Different types of algorithms for association rules
Implementation of different algorithms for association rules
Sequence learning using SPADE

The power of association is stronger than the power of beauty; therefore, the power of association is the power of beauty.
—John Ruskin

Congratulations on finishing the first part of the book! You explored the basics of unsupervised learning and algorithms like k-means clustering, hierarchical clustering, DBSCAN, principal component analysis, and others. It is expected that you have covered the mathematical concepts in the first part and created the Python codes to solve the exercise given at the end of each chapter.

Welcome to the second part of the book where we use the concepts learned in the first part and explore slightly more complex topics. We start with association rules in this chapter.

4.1 Technical toolkit

4.2 Association rule overview

4.3 The building blocks of association rules

4.3.1 Support, confidence, lift, and conviction

4.4 Apriori algorithm

4.4.1 Python implementation

4.4.2 Challenges with the Apriori algorithm

4.5 Equivalence class clustering and bottom-up lattice traversal

4.5.1 Python implementation

4.6 F-P algorithm

4.7 Sequence rule mining

4.7.1 Sequential Pattern Discovery Using Equivalence

4.8 Case study for association rules

4.9 Concluding thoughts

4.10 Practical next steps and suggested readings

Summary