We will go over several methods of determining patterns in data:
|
Type of model |
How the model works |
|---|---|
|
|
This model is used for itemset pattern detection, often used for shopping carts |
|
|
This model determines the co-occurrence of items in a dataset |
|
|
This model learns the association rules in a dataset |
|
|
This is an R package for mining sequences |
The Eclat algorithm is used for frequent itemset mining. In this case, we are looking for similar patterns in behavior, as opposed to looking for irregularities (like we did in other data mining approaches).
This algorithm uses intersections in the data to compute the support of candidates for events that frequently occur together, such as shopping cart items. The frequent candidates are then tested to confirm the pattern in the dataset.