ml.arules: Mining Association rules and Frequent Itemsets
In rfml: MarkLogic NoSQL Database Server in-Database Analytics for R

Description Usage Arguments Details Value

Mine frequent itemsets or association rules using MarkLogic Server built in Range Index functions. The function require that there is a Range Index on the underlying field of itemField, a range indexe can be created with the ml.add.index function. It will return a object that is of class rules or itemsets as defined in the arules package. It will need the arules package installed.

1 2	ml.arules(data, itemField, support = 0.5, confidence = 0.8, maxlen = 5, target = "rules")

`data`	an ml.data.frame object
`itemField`	a ml.data.frame field which is the field that the itemsets will be created of. The underlying field needs to have a Range Index defined.
`support`	a numeric value for the minimal support of an item set (default: 0.5)
`confidence`	a numeric value for the minimal confidence of rules/association hyperedges (default: 0.8)
`maxlen`	an integer value for the maximal number of items per item set (default: 5)
`target`	a character string indicating the type of association mined. One of "frequent itemsets" or "rules", default is "rules"

The frequent itemset and association rules extraction method is using the same method as the Apriori algorithm by first identify all 1-n itemsets that satisfy the support threshold and based on these extract rules that satisfy the confidence threshold.

It is depended on that there are a Range Index on the underlying field for the itemField. Information about the name of the field can be shown by mlDataFrame$itemField, where mlDataFrame is a ml.data.frame object and itemField is the name of the field.

Returns an object of class rules or itemsets.

rfml documentation built on May 2, 2019, 3:01 a.m.