R: aggregate dataframe but different column

Question

My data looks like this:

browsers.id browsers.label browsers.count browsers.id.1 browsers.label.1 browsers.count.1 
OS;FBSV        OS;FBSV            6        Chrome           Chrome                3 
Chrome         Chrome             86       Safari           Safari                47 
Chrome         Chrome             21       OS;FBSV          OS;FBSV               14

I want to aggregate or merge data to:

browsers.id count
Chrome       110
OS;FBSV      20
Safari       47

I try to use aggregate function,

aggregate(data[ ,3], list(data$browser.id), sum)

But it occurs error:

"arguments must have same length"

Could I fix it? Thanks.

akrun · Accepted Answer · 2015-11-28 05:24:47Z

1

We subset the 'id', and 'count' columns, convert the 'data.frame' to 'data.table' (setDT(..), reshape the 'wide' format to 'long' using melt (the data.table melt can take multiple measure columns), grouped by 'browsers.id', we get the sum of 'count'.

library(data.table) 
 melt(setDT(df1[c(1,3,4,6)]), measure=patterns('id', 'count'), 
   value.name=c('browsers.id','count'))[,list(count=sum(count)) ,
         by = browsers.id]
 #   browsers.id count
 #1:     OS;FBSV    20
 #2:      Chrome   110
 #3:      Safari    47

Or another option would be to rbind the 'id', 'count' columns

  rbindlist(list(df1[c(1,3)], df1[c(4,6)]))[,  
           list(count= sum(browsers.count)) ,browsers.id]

edited Nov 28, 2015 at 5:24

answered Nov 28, 2015 at 5:16

akrun

891k38 gold badges590 silver badges700 bronze badges

Sign up to request clarification or add additional context in comments.

1 Comment

Jeffery Chen Over a year ago

Thanks! Because I have more than six column, I try "setDT(df1)" and it works :)

Collectives™ on Stack Overflow

R: aggregate dataframe but different column

1 Answer 1

1 Comment

Your Answer

Hot Network Questions

Collectives™ on Stack Overflow

1 Answer 1

1 Comment

Your Answer

Sign up or log in

Post as a guest

Related