Comparison of algorithms of construction of associative rules on the basis of the data set of customer transactions

The article discusses the algorithms for constructing the association rules Apriori and Eclat, which are used to analyze a data set containing information about the grocery purchases of users of the largest US retailer Walmart. In the course of work, it is possible to obtain trivial and useful rules that can be taken into account when forming store departments and arranging goods in such a way as to increase consumer activity. The resulting graphs allow you to visually evaluate the constructed rules and make the most accurate predictions. In addition, the article compares two algorithms for finding associative rules for such parameters as changing the value of the support level and submitting a different amount of data to the input.