cannot make R leave enough room for labels in silhouette plot ('cluster' package)

gianmarco

TS Contributor
#1
Dear All,
I have problems making R plot the objects' labels in the 'silhouette' plot (out from the 'cluster' package).

See the attached image.
As you can see, the labels showing up in the left-hand side of the chart are truncated, i.e. there is not enough room to show them entirely. I went through the help documentation, but I couldn't find a viable solution.

Do you think it is possible to leave enough room to the left of the chart to be able to display long labels?

The following is the code for a sample dataset, and to get the attached chart. As for the second code, I have put some #annotations to explain what's going on.


Thank you.
Best

data:
Code:
mydata <- structure(c(0.47812386930075, 0.127251275299247, 0.0825022807431319, 
-0.390445104325078, -0.0319105644397104, 0.0764012513828784, 
0.17989154684742, 0.0376443260581566, -0.327365399920222, 0.315552204192887, 
-0.117494744401698, 0.0127663287534032, -0.178694714940958, 0.124638408742502, 
0.106751129148983, 0.0722788856574681, 0.173388026639424, 0.0498222975683006, 
0.139289827320685, -0.292159660786714, 0.302569367939057, -0.454996212422005, 
0.0733533758225443, 0.102283081186412, 0.0269966034008835, -0.291712006383772, 
-0.10965626974681, -0.0385013609564718, 0.0141619660722275, -0.0613162683720088, 
0.390864464926791, 0.032537394707918, -0.0981147624789522, 0.0727104299821822, 
0.0240118483102508, -0.0877489418588905, -0.15171595339673, 0.0423707720674034, 
0.0645146710970157, 0.108688258121981, 0.107329599470356, -0.0414354064004436, 
-0.129054801985255, 0.107189500792259, -0.17568816523869, 0.107565034778417, 
-0.0966435931474396, 0.0327278817962036, 0.019299439398085, -0.018178824214317, 
-0.0349129066691172, 0.0969316887880894, -0.0443427503497396, 
0.0694497769074093, 0.033032962969166, -0.0568557867350238, -0.0633999394784391, 
-0.0317823708665944, 0.0288019998764433, 0.039340224635988), .Dim = c(15L, 
4L), .Dimnames = list(c("A", "B", "C", "D", "E", "Geology", "Biochemistry", 
"Chemistry", "Zoology", "Physics", "Engineering", "Microbiology", 
"Botany", "Statistics", "Mathematics"), c("Dim 1", "Dim 2", "Dim 3", 
"Dim 4")))
code:
Code:
d <- dist(mydata, method = "euclidean") # calculate the distance matrix

fit <- hclust(d, method="ward.D2") # perform the hierc agglomer clustering

silhouette.data <- silhouette(cutree(fit, k=3),d) # store the silhouette data for the selected cluster solution (k=3)

row.names(silhouette.data) <- row.names(mydata) # give the objects' names to the rows of the object created in the above step

plot(silhouette.data)