Eclat

Intuition

Eclat is a very simple version of association "rule" learning that only uses support.
The output of the algorithm is the frequent itemsets.
- i.e. The sets of products that are frequently bought together and that are above a given min_support
- It is like truncating at Step 2 in the apriori algorithm.
- e.g. an output will look like: {eggs, milk, chicken}. Note that this is NOT a rule strictly speaking.
- The output tends to be obvious and it is heavily influenced by the high frequency items.
  - e.g. If bread, coffee and butter are very high frequency items, many of the frequent itemsets will be subsets of those 3 or contain one of them.
  - Output needs to be manually checked to identify some unexpected frequent itemsets.

See the apriori algorithm cheat sheet for the definition of support.

Steps

Step 1: Set a minimum support.
Step 2: Find the Frequent Itemsets. i.e. all the itemsets having support > min_support.
Step 3: Sort Frequent Itemsets by decreasing support.

Code

The Udemy couse did not cover Eclat for python. Google it!