Dear All,
my problem case: I want to use the apriori-function to analyze my test data.
Take a look (the data is attached).
WarenGruppe is a hierarchical higher class for Item. I analyzed with apriori()
Customer and Item but I want also to start a cross-rule-mining. That means for
example that one rule could be {Item2, Item4} -> {WG2} and another one could be
{Item1, Item5} -> {Item2}.
The code is attached. I can just analyze WarenGruppe or Item, but the goal is
to analyze them booth (like the example).
Has somebody a hint or a reason to realize this "cross-Analysis" in R?
Best regard
Michael
Customer Item WarenGruppe
2 item2 WG2
2 item3 WG3
3 item1 WG1
3 item2 WG2
3 item4 WG2
4 item1 WG1
4 item3 WG3
5 item2 WG2
5 item3 WG3
6 item1 WG1
6 item3 WG3
9 item1 WG1
9 item2 WG2
9 item5 WG3
1 item2 WG2
1 item4 WG2
7 item1 WG1
7 item2 WG2
7 item3 WG3
7 item5 WG3
8 item1 WG1
8 item2 WG2
8 item3 WG3
### Analyse with apriori the table Customer_item
# Data import
item_testdata <-
read.table("C:/Users/michael.fueg/Documents/R/Daten/Customer_item.txt",header=TRUE)
WarenGruppe=c("WG1","WG2","WG3","WG2","WG3") # We have Item1,...Item5
# Build transaction-class
library(arules)
trans <- as(split(item_testdata[,"Item"], item_testdata[,"Customer"]),
"transactions")
# Build hierarchical variable
itemInfo(trans)[["WarenGruppe"]] <- WarenGruppe
summary(trans)
inspect(trans)
trans.aggregate <- aggregate(trans, WarenGruppe)
inspect(trans.aggregate)
# Call apriori
rules <- apriori(trans)
inspect(rules)
rules.aggregate <- aggregate(rules, WarenGruppe)
inspect(rules.aggregate)
______________________________________________
R-help@r-project.org mailing list
https://stat.ethz.ch/mailman/listinfo/r-help
PLEASE do read the posting guide http://www.R-project.org/posting-guide.html
and provide commented, minimal, self-contained, reproducible code.
