Suppose, we have 3 people called: Francis, Cedric and Nina. Base on what
they have eaten, we want to cluster people by "diet",
"non-diet".
# original data file, named as filename "food.csv".
Francis|potato
Francis|chocolate
Francis|chocolate
Francis|milk
Cedric|vegetable
Cedric|vegetable
Cedric|potato
Nina|potato
Nina|chocolate
Nina|chocolate
Nina|potato
# Step 1: I plan to reform the food.csv file into following frame:
Francis Cedric Nina
potato 1 1 2
chocolate 2 0 2
milk 1 0 0 0
fish 0 1 0 0
vegetable 0 2 0
But, I have problem to reform the original file into that way I want.
Basically, I have only tried:>
org<-read.table("food.csv",col.name=c("name","food_list"),
sep="|") #
import file to R> x<-data.frame(Francis=rep(0,times=5)) # set up the frame of file
> rownames(x)<-org$food_list
here I failed, and don't know how to continue.
Please help me out with the proper R command.
# Step 2: I want to draw 3 overlapped barplot with x-axis(5 food name list),
y-axis(the occurence value of each food by each person).
# Step 3: By comparing the barplots, I should be able to draw the
conclusion: Francis and Nina are in the group of "non-diet", and
Cedric is
in the group of "diet".
Well, that's my simple strategy of "clustering", if any of you
have better
suggestion, I really appreciate it.
Thanks!
[[alternative HTML version deleted]]
