Sunday, 18 August 2013

Categorize, group and clustering data

Categorize, group and clustering data

pIf I would order the data from Titanic like the following output/p
precodedflt;-data.frame(Titanic) df_Crew lt;- df[df$Class==Crew,] L lt;-
lapply(1:4, function(i) aggregate(df_Crew$Freq, by=df_Crew[1:i], sum)) L2
lt;- lapply(L, function(d) data.frame(group=do.call(paste,
c(as.list(d[names(d)!=x]), sep=_)), freq=d$x)) L3lt;-data.frame() for(i in
1:3){ dlt;-cbind(from=rbind(L2[[i]],L2[[i]])$group,L2[[i+1]])
L3lt;-rbind(L3,d) } L3 from group freq 1 Crew Crew_Male 862 2 Crew
Crew_Female 23 3 Crew_Male Crew_Male_Child 0 4 Crew_Female
Crew_Female_Child 0 5 Crew_Male Crew_Male_Adult 862 6 Crew_Female
Crew_Female_Adult 23 7 Crew_Male_Child Crew_Male_Child_No 0 8
Crew_Female_Child Crew_Female_Child_No 0 9 Crew_Male_Adult
Crew_Male_Adult_No 670 10 Crew_Female_Adult Crew_Female_Adult_No 3 11
Crew_Male_Child Crew_Male_Child_Yes 0 12 Crew_Female_Child
Crew_Female_Child_Yes 0 13 Crew_Male_Adult Crew_Male_Adult_Yes 192 14
Crew_Female_Adult Crew_Female_Adult_Yes 20 /code/pre pThen I could greate
a nice tree like/p precodelibrary(igraph) g lt;- graph.data.frame(L3,
directed=TRUE)
plot(g,layout=layout.reingold.tilford(g,root=1),edge.arrow.size=0.5)
/code/pre pIf I would set a better layout the tree would become a very
nice look and feel./p pWhat I have done, I cluster the data Titanic and
print a graph. I can very well imagine there already exists pkg they are
designed for group data in several ways, depending on the question the
R-user impose to the data. So this pkg contains general functions for
clustering and plotting data also in the way I do above. strongSo would
you please sorted out for me wich function can do this e.g. with the data
titanic?/strong /p

No comments:

Post a Comment