I'm applying a Market Basket Analysis using R script. I have a big data set in the following format:
1 25 20 21 36
2 45 23 17
3 16 18 26 13 14
4 37 41
5 15 17 25
Where the first column is the ID of the Transaction and the others the products purchased.
I want to create a link analysis to identify which products are correlated. I've this R script:
library(arules)
library(arulesViz)
library(igraph)
library(iplots)
data <- read.transactions('data.csv', sep=',')
itemFrequencyPlot(data,
type="absolute",
topN=10,
horiz=TRUE,
col='steelblue3',
xlab=',',
main='Item frequency, relative')
itemsets = apriori(data, list(supp=0.05,conf=0.3,minlen=2))
plot(itemsets, method="graph",
control=list(nodeCol="steelblue3", edgeCol="black", type="items"), shading="lift")
I'm getting the following images:
But what I'm trying to get is a map that contains all the products and based on the line that connect the nodes I can see the confidence between the products (I don't want the balls, I need lines).
Do you know how can I do that?