On Sat, Mar 12, 2011 at 10:02 AM, Graham Williams <graham.willi...@togaware.com> wrote: > On 12 March 2011 00:07, Xiaobo Gu <guxiaobo1...@gmail.com> wrote: >> On Fri, Mar 11, 2011 at 2:55 AM, Graham Williams >> <graham.willi...@togaware.com> wrote: >>> Did you scroll down the window to see the rules? >> OK, it takes a long time for rattle to show the rules, about 30 >> seconds, and why the message on the status bar is "the decision tree >> model has been built. Time taken:0.01 secs", I have attached the >> little dataset used to do the test. > > I've fixed the incorrect "decision tree" text. Thanks. > > For the other issue see below. And thanks for supplying a repeatable > sample and dataset. > >> Another question, why the confidence of the rules are all 1 ? > >>> On 10 March 2011 02:07, Xiaobo Gu <guxiaobo1...@gmail.com> wrote: > >>>> set transactions ...[35 item(s), 8 transaction(s)] done [0.00s]. > > That does not look right?
I think it's because there are to few sample records, so all the rules are with 100% confidence > Be sure to make "items" your Target and > "tx_no" your Ident rather than the other way around. How to save the rules and use the rules to do product recommandations with Rattle? > > Regards, > Graham > ______________________________________________ R-help@r-project.org mailing list 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.
