Association Rule Learning - Achronus/Machine-Learning-101 GitHub Wiki

'People who bought... also bought...'

Association rules analysis is a technique to uncover how items are associated to each other.

Table of Contents

Apriori

This algorithm is commonly used in data mining. It consists of 3 parts - support, confidence and lift.

Apriori

Using a list of movies as an example, M stands for a set of movies.

Support

Confidence

Lift

This is how it works:

  • Step 1 - Set a minimum support and confidence.
  • Step 2 - Take all subsets in transactions having higher support than minimum support.
  • Step 3 - Take all the rules of these subsets having higher confidence than minimum confidence.
  • Step 4 - Sort the rules by decreasing lift.

See the code here for an example of an Apriori model. This uses an amazing python script apyori from ymoch.

# Training Apriori on the dataset
from apyori import apriori
rules = apriori(transactions, min_support = 0.003, min_confidence = 0.2, min_lift = 3, min_length = 2)

Eclat

This is a simplified version of Apriori and is used to identify basics information on sets of items that have been purchased together. This only has 1 part - the support.

Support

This is how it works:

  • Step 1 - Set a minimum support.
  • Step 2 - Take all the subsets in transactions having higher support than minimum support.
  • Step 3 - Sort these subsets by decreasing support.