Michael, thanks a lot, it works!
Now I have to study the dataset to mining some interesting rules.
Just one more question, I saw ways to find rules that contains some itens.
But, Is there a method to find rules that doesn't have a item?
Thanks again! You were very helpfull!!!
2009/12/4 Michael Hahsler hahs...@ai.wu.ac.at
Alexandre,
You are mining for association rules with an absolute support count of
1001*0.01 = 10.1. From your 40 minutes processing time I guess that your
survey data set is very dense and that results in an enormous number of
frequent itemsets (potentially up to 2^k - k - 1 which is in your case for
k=71 about 10^21) which causes you to run out of memory. You have the
following options:
* increase minimum support (e.g., start with 0.1) and see how low you can
go without using up all your memory (I don't know how to watch memory usage
on Windows).
* restrict the maximal length of frequent itemsets by using
parameter=list(maxlen = 3, support = ...)
If that does not help I will need your dataset and some code to reproduce
and study the problem.
-Michael
Hi I'm a windows XP user. My notebook have 1gb ram, 160gb hd, processor
amd turion64 1,6gh. For processing, it takes about 40 minutes. This is the
code i used: dados=read.csv(C:/Documents and Settings/Administrador/Meus
documentos/My Dropbox/Estat?stica/Association Rules/Top2009 alterado.csv,
header=T, sep=;) library(arules) bin=as(dados, transactions) rules -
apriori(bin, parameter = list(support = 0.01, confidence = 0.6)) Below is a
sample of the file. I export data as CSV from excel. I have 71 columns
(variabels) an 1001 lines (responses) 1. Churrascaria 2. Supermercado 3.
Restaurante Self Service 4. Restaurante Chin?s 5. Buffet 6. Sorvete Galp?o
Nelore Super Muffato N?o Sabe Jin Jin Planalto S?vio Vento Sul Super Muffato
N?o Sabe N?o Sabe N?o Sabe Doce Ver?o Ga?cha Super Muffato N?o Sabe N?o Sabe
N?o Sabe Kibon Tradi??o Ga?cha Super Muffato N?o Sabe N?o Sabe N?o Sabe
Nestl? N?o Sabe Super Muffato N?o Sabe N?o Sabe Estilo S?vio Rancho Grill
Viscardi Akira Akira N?
o Sabe N?o Sabe Thank you very for your helping!!! On 3 dez, 01:46, Steve
Lianoglou mailinglist.honey...@gmail.com wrote:
Hi,
On Wed, Dec 2, 2009 at 6:57 PM, Alexandre - UEL shima...@gmail.com
wrote:
Hello everybody!
I'm trying some datamining, but i'm having some problems with arule
package, in the end of processing R had to be closed. I already
tryied to reinstall the 2.10 version, change the computer and
realocated more virtual memory.
Does anyone had this problem to?
I had a hiphoteses that i have to prepare the data, somehow i don't
know.
Thanks for helping!!!
Can you provide more info here?
1. I'm assuming since you're talking about reallocating virtual
memory, or whatever, you're on windows?
2. What's the exact error you're getting (what's it saying before R
'had to be closed'?
3. What's the size of your data? What type of data is it?
4. How much RAM do you have?
5. Are you on a 32 or 64 bit system?
6 What happens if you cut your data in half?
6. Can you provide a (very small) reproducible example of your data +
code?
...
-steve
--
Steve Lianoglou
Graduate Student: Computational Systems Biology
| Memorial Sloan-Kettering Cancer Center
| Weill Medical College of Cornell University
Contact
Info:http://cbio.mskcc.org/~lianos/contacthttp://cbio.mskcc.org/%7Elianos/contact
__
r-h...@r-project.org mailing listhttps://
stat.ethz.ch/mailman/listinfo/r-help
PLEASE do read the posting guidehttp://
www.R-project.org/posting-guide.html
and provide commented, minimal, self-contained, reproducible code.
--
Michael Hahsler
email: mich...@hahsler.net
web: http://michael.hahsler.net
--
É inútil e estúpido lastimar o mundo.
Sofrimento e prazer são frutos do que faz
Nissen Shounin
[[alternative HTML version deleted]]
__
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.
