R help - Jun 2012 - Arules - predict function issues - subscript out of bounds

Hi,

I'm doing a Market Basket Analysis in which I have a list of transaction
id's in column 2 and transactions(product names) in column 1. I read this
into a transaction file using a
txn<-read.transaction(file="data.csv",format='single',
rm.duplicates=F,
cols=c(1,2))

If I want to use the apriori algorithm everything seems to be running fine.
However it is when I want to form cluster of these transactional patterns I
find problems. I formed clusters using the following code:

s<-sample(txn,1000)
d<-dissimilarity(s, method="Jaccard")
clustering<-pam(d,k=5)

But when I'm trying to predict this on the larger set it keeps throwing an
subscript out of bound error

Label<-predict(s[clustering$mediods],txn,method="Jaccard")

Can anyone explain to me why this keeps happening ?? I've tried this on
other datasets like Groceries/ Adult in the arules package and it seems to
work fine !!

Thanks,
Ankur

	[[alternative HTML version deleted]]

Hi Ankur, 

I am running into the exact same issue you have described above. Were you
able to find out why it didn't work on your data set and resolve it? If yes,
could you share? 

Much thanks & regards,
Alice 



--
View this message in context:
http://r.789695.n4.nabble.com/Arules-predict-function-issues-subscript-out-of-bounds-tp4634422p4644546.html
Sent from the R help mailing list archive at Nabble.com.