Need help with R codes and visualizations for the questions below. these are que
ID: 3206443 • Letter: N
Question
Need help with R codes and visualizations for the questions below. these are questions from a textbook and the datasets are part of the installed packages in R and can be installed directly in R.
Q6: Wine
The wine dataset can be found in the packages gclus, MMST, pgmm, and rattle. They took the data from the UCI Machine Learning Repository [Bache and Lichman, 2013]. The original source is an Italian software package [Forina et al., 1988]. The version in pgmm has about twice as many variables as the others, and the version in MMST includes the names of the three classes of wine, rather than the numeric coding that the other versions use.
a) Use pcp's to investigate how well the variables separate these classes.
b) Are there any outliers?
c) Is there evidence of subgroups within the classes?
Explanation / Answer
There is direct command for pcp in r this as fallows
commands:
data(wine,package="gclus")
ggparcoord(data=wine,columns=c(2:12),scale=uniminmax)
if the command show the error then load package namely "GGally"
## for more detail about the command take help from R
In gclus pckage it give variable and nane of the class of the wine.
from the graph we may conclude about the outlier.