Apriori Algorithm

The Apriori algorithm is a data mining methodology for identifying and quantifying frequent item sets in transaction data, and is the foundational component of association rule learning. Extending the results of the Apriori algorithm to association rule learning will be discussed in the next section. The minimum value to qualify as frequent in the Apriori algorithm is an input into the model and, as such, is adjustable. Frequency is quantified here as support, so the value inputted into the model is the minimum support acceptable for the analysis being done. The model then identifies all item sets whose support is greater than, or equal to, the minimum support provided to the model. Note that the minimum support parameter is not a parameter that can be optimized via a grid search because there is no evaluation metric for the Apriori algorithm. Instead, the minimum support parameter is set based on the data, the use case, and domain expertise.

The main idea behind the Apriori...