I have a data set of Zip code and house code.
df = data.frame(zip = c(2900,2900,2900,3200,3100,3200),
house_code = c('abc','cde','efg','ghi','ijk','klm'))
I need to find top 2 zip code in terms of number of house_code?
Asked
Active
Viewed 80 times
-1
codekiller
- 53
- 1
- 9
-
you forgot quotes around your strings...my edits are full or else I'd edit it for you. – Cyrus Mohammadian Aug 30 '16 at 04:48
-
@CyrusMohammadian: Thanks :) – codekiller Aug 30 '16 at 04:49
-
3also, your question isn't clear (at least to me). what does "in terms of number of `house_code` mean"? your `house_code` isnt numeric so you cant mean in terms of which has the highest number and they also dont repeat (in your given example), so you must not mean by count either. So what do you mean? – Cyrus Mohammadian Aug 30 '16 at 04:49
-
1`dplyr::count(df, zip, sort = TRUE)[['zip']][1:2]` – alistaire Aug 30 '16 at 05:39
2 Answers
0
I think it could be: head(df[df$house_code == 'some value']$zip,2) where 'some value' is a house_code entry.
Josseline Perdomo
- 363
- 1
- 4
- 15
0
First use table to match house_codes and zip_codes.
> dftable <- table(df)
house_code
zip abc cde efg ghi ijk klm
2900 1 1 1 0 0 0
3100 0 0 0 0 1 0
3200 0 0 0 1 0 1
Then use rowSums to find the number of house_codes for each zip_code.
> numHouse <- rowSums(dftable)
2900 3100 3200
3 1 2
Finally use order to find the top 2.
> names(numHouse)[order(numHouse, decreasing = TRUE)[1:2]]
[1] "2900" "3200"
Barker
- 2,074
- 2
- 17
- 31
-
1Just call `table(df$zip)`, i.e. `names(sort(table(df$zip), TRUE)[1:2])` – alistaire Aug 30 '16 at 05:36