R code to do some affinity analysis on ClickStream data. Requires a three packages (arules
, arulesViz
, Rgraphviz
) which are not found in base R package; I don't use require
because I've already installed them.
In R terminal, source
the file and then use the provided functions on the transaction data set. This data set must be in the basket
format, where each transaction contains comma-separated values of items, e.g.,
bread, fruit, milk
fruit, milk
It also requires Christian Borgelt's fpgrowth
code to be installed on the computer.