Hi, I have 100000 units with 10 attributes (attr1, attr2, attr3, etc...) For instance: unit attr1 attr2 attr3 ... 1 a ww 12 2 a re 11 3 b ww 09 4 c yt 02 5 a qw 02 ... I'd like to answer to the question: a) what are the most frequent combinations of attributes? b) How could I describe the relations among the attributes? c) What are the most significative values for each attribute and how they are in relationship with the value of others attributes? Do you suggest any specific method in order to answer to these questions? Thanks -- View this message in context: http://r.789695.n4.nabble.com/Knowledge-discovery-tp2276207p2276207.html Sent from the R help mailing list archive at Nabble.com.
Hello, For the first question, you can use "table" function. Read how to apply it, it is very straitghforward. Once you have the most common combination, you have to investigate further this combination to apply a statistical model to it. What model?. It depends on what are you looking for. Regards, Carlos. On Fri, Jul 2, 2010 at 11:37 AM, abanero <gdevitis@xtel.it> wrote:> > Hi, > > I have 100000 units with 10 attributes (attr1, attr2, attr3, etc...) > > For instance: > > unit attr1 attr2 attr3 ... > > 1 a ww 12 > 2 a re 11 > 3 b ww 09 > 4 c yt 02 > 5 a qw 02 > ... > > I'd like to answer to the question: > > a) what are the most frequent combinations of attributes? > b) How could I describe the relations among the attributes? > c) What are the most significative values for each attribute and how they > are in relationship with the value of others attributes? > > Do you suggest any specific method in order to answer to these questions? > > Thanks > > > -- > View this message in context: > http://r.789695.n4.nabble.com/Knowledge-discovery-tp2276207p2276207.html > Sent from the R help mailing list archive at Nabble.com. > > ______________________________________________ > 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. >[[alternative HTML version deleted]]
with "table" function you can just build a contigence table. What do you think about "arules" package? I thought "mining associative rules" is the correct approach to the problem.. Thanks Abanero -- View this message in context: http://r.789695.n4.nabble.com/Knowledge-discovery-tp2276207p2276368.html Sent from the R help mailing list archive at Nabble.com.
On Jul 2, 2010, at 5:37 AM, abanero wrote:> > Hi, > > I have 100000 units with 10 attributes (attr1, attr2, attr3, etc...) > > For instance: > > unit attr1 attr2 attr3 ... > > 1 a ww 12 > 2 a re 11 > 3 b ww 09 > 4 c yt 02 > 5 a qw 02 > ... > > I'd like to answer to the question: > > a) what are the most frequent combinations of attributes? > b) How could I describe the relations among the attributes? > c) What are the most significative values for each attribute and how > they > are in relationship with the value of others attributes? > > Do you suggest any specific method in order to answer to these > questions?RSiteSearch("visualizing categorical data") -- David Winsemius, MD West Hartford, CT