[R] How can you validate association rules?

Jeff Newmiller jdnewmil at dcn.davis.CA.us
Sat Feb 14 17:21:17 CET 2015


It will help greatly if you can learn to communicate according to the norms of the list. Reading the Posting Guide mentioned at the bottom of this email will help.

1) This mailing list is plain text. If you send HTML email it will get converted and we won't see what you see, which leads to  miscommunication.

2) The topic here is R, not statistics, so be careful not to wander too far off into why algorithms work. There are other forums where that kind of discussion makes sense, such as stats.stackexchange.com. Your question in its current form seems to belong there, though it sounds like you could probably get on topic here with a little work. Have you read the CRAN task view on machine learning? [1]

3) A small, reproducible example expressed in R of what you have achieved is highly recommended [2]. A clear specification of what you want to achieve is also needed. (We could probably make  progress on this question here if you had the example.) If you reference examples in R help files that you have read then you will clarify how far along you are in understanding the use of R.

[1] http://cran.r-project.org/web/views/MachineLearning.html

[2] http://stackoverflow.com/questions/5963269/how-to-make-a-great-r-reproducible-example
---------------------------------------------------------------------------
Jeff Newmiller                        The     .....       .....  Go Live...
DCN:<jdnewmil at dcn.davis.ca.us>        Basics: ##.#.       ##.#.  Live Go...
                                      Live:   OO#.. Dead: OO#..  Playing
Research Engineer (Solar/Batteries            O.O#.       #.O#.  with
/Software/Embedded Controllers)               .OO#.       .OO#.  rocks...1k
--------------------------------------------------------------------------- 
Sent from my phone. Please excuse my brevity.

On February 13, 2015 8:46:32 AM PST, ayane <celestialxangel at hotmail.com> wrote:
>Hey everyone, 
>
>I just started learning R and statistics. I made association rules with
>the apriori algorithm in the arules package. I sorted them and pruned
>them. But now I want to validate/test them. In supervised learning they
>use cross validation. In this paper by Patrick O Perry
>http://arxiv.org/abs/0909.3052 I read it is possible to use cross
>validation on unsupervised learning methods (such as arules). But since
>my background in R, statistics and machine learning is weak the paper
>was quite hard to get through (also I'm not in college, I just finished
>high school and am preparing to study computer sciencec). 
>
>So my question is, how do you validate your rules? How can you use
>cross validation to do so? I have my main data set "Orders", it has 2
>million records. I used the sample function to make a training data set
>and a test data set. But what do I do next? Are there other techniques
>I should know of?
>
>If there are any specific tutorials on this subject, I would also love
>to learn from them.
>
>Kind regards,
>Ayane 		 	   		  
>	[[alternative HTML version deleted]]
>
>______________________________________________
>R-help at r-project.org mailing list -- To UNSUBSCRIBE and more, see
>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.



More information about the R-help mailing list