0

I am using the apriori algorithm in r to find frequent itemsets. Now I want to know if a sample will produce the same frequent itemsets as the whole data set will, and I was wondering if there exists a way to compare these two. 

    itemset = apriori(allTransactions, parameter = list(support = 0.01, target="frequent itemsets", maxlen = 100));

and I want to compare this (accuracy, recall, precision etc) to 

    itemsetSample = apriori(sampleTransactions, parameter = list(support = 0.01, target="frequent itemsets", maxlen = 100));

Thanks!

Maria
  • 13
  • 3
  • You may want to ask this question at another community. [Data Science Community](http://datascience.stackexchange.com/questions/9705/calculate-the-overall-accuracy-of-mined-rules-using-apiriori-algorithm?newreg=363e15981c9f4a7d9779cc4331cab0e4) already has this question but it is unanswered. Moreover, during my practice of data mining, for association rules measure like lift, support and confidence are more useful (which the package provide by itself). – M-- Mar 31 '17 at 14:07
  • Another comment on the question here. Please provide a reproducible code so you can get a better help. `dput` can be a good tool or you can refer to an online dataset for this particular example because it needs sufficient number of records. read this [guide](http://stackoverflow.com/questions/5963269/how-to-make-a-great-r-reproducible-example#answer-5963610) for more information. – M-- Mar 31 '17 at 14:13

0 Answers0