Grouping table by column value in R

Question

I'm sorry for the newbie question but I cannot find an answer.

I have a table with 3 columns:

File  Word  Occurrences
  f1   cat            2
  f1   dog            1
  f2   cat            4
  f2   pig            3

And I want to convert it in a table where each file corresponds to a row and each column to the occurrences of a specific word:

File  Cat  Dog  Pig
  f1    2    1    0
  f2    4    0    3

@akrun http://stackoverflow.com/a/5890831/3710546, from first link. — , Nov 11 '15 at 12:52
@Pascal Thanks for that url. But as mentioned [here](http://stackoverflow.com/questions/33636641/combine-multiple-rows-with-same-field-in-r/33636750#33636750), it is not a valid reason to close the question (this link was also closed and then reopened based on that). — akrun, Nov 11 '15 at 12:55
@akrun It is your viewpoint that it is not a dupe. It is for me. That's all. — , Nov 11 '15 at 12:56
@Pascal I didn't say it is your viewpoint. But, you can close it. I will undupe it. :-) — akrun, Nov 11 '15 at 12:57
@Pascal It is also not my viewpoint. It is based on the precedence in that link. — akrun, Nov 11 '15 at 12:58
My point is that if another clear dupe can be unduped, this can be too. — akrun, Nov 11 '15 at 13:03
@akrun, for what reason? Just because a mistake was possibly made somewhere else (by "unduping" a duplicate), doesn't mean it should be repeated elsewhere. That's at least my opinion.. — talat, Nov 11 '15 at 13:06
@akrun, I don't see anything put forward by Ananda here and I don't know what other question you referred to. My comment was general — talat, Nov 11 '15 at 13:08
@docendodiscimus It is there in the discussion `@Frank, would you know that xtabs sums without either trying it out or trying to interpret the help page for xtabs?` — akrun, Nov 11 '15 at 15:25

score 3 · Accepted Answer · answered Nov 11 '15 at 12:39

3

Try

library(reshape2)
 dcast(df1, File~Word, value.var='Occurrences', sum)

Or

 xtabs(Occurrences~File+Word, df1)

answered Nov 11 '15 at 12:39

akrun

874,273
37
540
662

1

Thanks, both solution works! I have used the second – Alessandro Valentini Nov 11 '15 at 15:38

score 2 · Answer 2 · answered Nov 11 '15 at 13:07

2

my try without additional packages:

d <- read.table(header=TRUE, text=
'File  Word  Occurrences
f1   cat            2
f1   dog            1
f2   cat            4
f2   pig            3')
d.w <- reshape(d, dir="wide", idvar="File", timevar="Word")
d.w[is.na(d.w)] <- 0
d.w

answered Nov 11 '15 at 13:07

jogo

12,469
11
37
42

Thanks, even this is working! – Alessandro Valentini Nov 11 '15 at 15:39

Grouping table by column value in R

2 Answers2