I am using market basket analysis/apriori rules algorithm for the first time. My data looks like the following:
mdinc1 <- c(0,1,0)
mdinc2 <- c(1,0,0)
mdinc3 <- c(0,0,1)
mdnrent1 <- c(0,0,1)
mdnrent2 <- c(0,1,0)
mdnrent3 <- c(1,0,0)
mdnage1 <- c(0,0,1)
mdnage2 <- c(0,1,0)
mdnage3 <- c(1,0,0)
df <- data.frame(mdinc1, mdinc2, mdinc3, mdnrent1, mdnrent2, mdnrent3, mdnage1, mdnage2, mdnage3)
I am using the following code:
rules <- apriori(primary1, parameter=list(supp=0.3, conf=0.8))
I use a support of 0.3 because each item is a quantile so it should show up 1/3rd of the time. When I run the code, I get the following error:
Error in apriori(primary1, parameter = list(supp = 0.3, conf = 0.5)) :
not enough memory. Increase minimum support!
I feel like this is a fairly high minimum support. Should I reduce the number of variables or is there an alternative factor analysis/classification approach that is worth considering?
