Tom D. Harray
2016-Sep-12 14:39 UTC
[R] Arules Package: Rules subset with 'empty' left hand side (lhs)
Hello,
subsets of association rules (with respect to support, confidence, lift, or
items) can be obtained with the arules::subset() function; e.g.
rm(list = ls(all.names = TRUE))
library(arules)
set.seed(42)
x <- lapply(X = 1:500, FUN = function(i)
sample(x = 1:10, size = sample(1:5, 1), replace = FALSE)
)
x <- as(x, 'transactions')
rules <- apriori(
data = x,
parameter = list(target = 'rules', minlen = 1, maxlen = 2,
support = 0.10, confidence = 0.32)
)
rules <- arules::sort(x = rules, decreasing = TRUE, by ='support')
gives the rules
3 {} => {1} 0.330 0.3300000 1.0000000
2 {} => {3} 0.326 0.3260000 1.0000000
1 {} => {2} 0.320 0.3200000 1.0000000
20 {3} => {1} 0.120 0.3680982 1.1154490
21 {1} => {3} 0.120 0.3636364 1.1154490
16 {4} => {3} 0.114 0.3677419 1.1280427
(...)
However, I cannot figure out (help/web) how to get the subset for the rules
with empty left hand side (lhs) like subset(rules, lhs == ''). I could
run the
apriori() function twice and adjust the min/maxlen parameters as a band
aid fix.
So my question is: How do I subset() association rules with empty lhs?
Thanks and regards,
Dirk
Luisfo
2016-Sep-13 09:10 UTC
[R] Arules Package: Rules subset with 'empty' left hand side (lhs)
Dear Tom, I think this is the line you need arules::subset(rules, subset=lhs %pin% "") I found the solution here: http://stackoverflow.com/questions/27926131/how-to-get-items-for-both-lhs-and-rhs-for-only-specific-columns-in-arules One more thing. For printing the rules, I needed the inspect() command you didn't provide. I hope this helps. Best, *Luisfo Chiroque* /PhD Student | PhD Candidate IMDEA Networks Institute/ http://fourier.networks.imdea.org/people/~luis_nunez/ <http://fourier.networks.imdea.org/people/%7Eluis_nunez/> On 09/12/2016 04:39 PM, Tom D. Harray wrote:> Hello, > > subsets of association rules (with respect to support, confidence, lift, or > items) can be obtained with the arules::subset() function; e.g. > > rm(list = ls(all.names = TRUE)) > library(arules) > set.seed(42) > > x <- lapply(X = 1:500, FUN = function(i) > sample(x = 1:10, size = sample(1:5, 1), replace = FALSE) > ) > x <- as(x, 'transactions') > > rules <- apriori( > data = x, > parameter = list(target = 'rules', minlen = 1, maxlen = 2, > support = 0.10, confidence = 0.32) > ) > rules <- arules::sort(x = rules, decreasing = TRUE, by ='support') > > gives the rules > 3 {} => {1} 0.330 0.3300000 1.0000000 > 2 {} => {3} 0.326 0.3260000 1.0000000 > 1 {} => {2} 0.320 0.3200000 1.0000000 > 20 {3} => {1} 0.120 0.3680982 1.1154490 > 21 {1} => {3} 0.120 0.3636364 1.1154490 > 16 {4} => {3} 0.114 0.3677419 1.1280427 > (...) > > However, I cannot figure out (help/web) how to get the subset for the rules > with empty left hand side (lhs) like subset(rules, lhs == ''). I could run the > apriori() function twice and adjust the min/maxlen parameters as a band > aid fix. > > > So my question is: How do I subset() association rules with empty lhs? > > > Thanks and regards, > > Dirk > > ______________________________________________ > 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.[[alternative HTML version deleted]]
Tom D. Harray
2016-Sep-13 13:30 UTC
[R] Arules Package: Rules subset with 'empty' left hand side (lhs)
Hello Luisfo, thank you for the hint: Your suggestion arules::subset(rules, subset=lhs %pin% "") gave 18 rules (out of 21) in my example, and not 3, what I have expected. Surprisingly the negation of the subset condition arules::subset(x = rules, subset = !(lhs %pin% "")) returns the 3 rules with empty lhs. Hello Martin, I add you to this thread, because the arules::subset() behaviour appears to me to be a bug in arules. And I'd like to suggest to add an explanation/example to arules::subset() help. Cheers, Dirk On 13 September 2016 at 05:10, Luisfo <luisfo89 at yahoo.es> wrote:> Dear Tom, > > I think this is the line you need > arules::subset(rules, subset=lhs %pin% "") > I found the solution here: > http://stackoverflow.com/questions/27926131/how-to-get-items-for-both-lhs-and-rhs-for-only-specific-columns-in-arules > > One more thing. For printing the rules, I needed the inspect() command you > didn't provide. > > I hope this helps. > > Best, > > Luisfo Chiroque > PhD Student | PhD Candidate > IMDEA Networks Institute > http://fourier.networks.imdea.org/people/~luis_nunez/ > > On 09/12/2016 04:39 PM, Tom D. Harray wrote: > > Hello, > > subsets of association rules (with respect to support, confidence, lift, or > items) can be obtained with the arules::subset() function; e.g. > > rm(list = ls(all.names = TRUE)) > library(arules) > set.seed(42) > > x <- lapply(X = 1:500, FUN = function(i) > sample(x = 1:10, size = sample(1:5, 1), replace = FALSE) > ) > x <- as(x, 'transactions') > > rules <- apriori( > data = x, > parameter = list(target = 'rules', minlen = 1, maxlen = 2, > support = 0.10, confidence = 0.32) > ) > rules <- arules::sort(x = rules, decreasing = TRUE, by ='support') > > gives the rules > 3 {} => {1} 0.330 0.3300000 1.0000000 > 2 {} => {3} 0.326 0.3260000 1.0000000 > 1 {} => {2} 0.320 0.3200000 1.0000000 > 20 {3} => {1} 0.120 0.3680982 1.1154490 > 21 {1} => {3} 0.120 0.3636364 1.1154490 > 16 {4} => {3} 0.114 0.3677419 1.1280427 > (...) > > However, I cannot figure out (help/web) how to get the subset for the rules > with empty left hand side (lhs) like subset(rules, lhs == ''). I could run > the > apriori() function twice and adjust the min/maxlen parameters as a band > aid fix. > > > So my question is: How do I subset() association rules with empty lhs? > > > Thanks and regards, > > Dirk > > ______________________________________________ > 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. > >