New dataframe with sums based on conditions of another dataframe

Question

I have a question regarding dataframe manipulation. I have a dataframe which looks like this:

year | month | deviceCategoy  | sessions

2017 | 4     | desktop        | 140000

2017 | 4     | mobile         | 200000

2017 | 4     | tablet         | 80000

...   ...       ...             ...

All in all the dataframe contains data for the whole year 2017 and 2018 until now. Now I want to have a new dataframe, where I only have the device categories desktop and mobile. The sessions for the tablets should be added to desktop.

The result should look like this:

year | month | deviceCategoy  | sessions

2017 | 4     | desktop        | 220000

2017 | 4     | mobile         | 200000

Does anybody knows how to do this?

Answer 1

We could change the "tablet" strings to "desktop" and then do an aggregate

i1 <- df1$deviceCategoy == "tablet"
df1$deviceCategoy[i1] <- "desktop"
aggregate(sessions ~ ., df1, sum)
#   year month deviceCategoy sessions
#1 2017     4       desktop   220000
#2 2017     4        mobile   200000

---

Or using tidyverse

library(dplyr)
df1 %>%
   mutate(deviceCategoy = replace(deviceCategoy, deviceCategoy == "tablet", "desktop")) %>%
   group_by_at(names(.)[1:3]) %>% 
   summarise(sessions = sum(sessions))