June 2017
Beginner to intermediate
576 pages
15h 22m
English
We have seen in the data prep stage that there are a large number of itemsets generated for each invoice. To begin to demonstrate the algorithm, we will extract one representative word from each product description, and use that word as the consequent (or rhs) to build some association rules. We have already saved the first and last words from each product description. We would examine those words more closely and see if we can filter them to result in a manageable set of transactions.
Let's first preview the frequency of the first and last word of the descriptions in descending order. That should give us a clue as to what the popular products are:
library(arules) > Loading required package: Matrix > > Attaching ...