CODEWITHSUNDEEP
rdplyr
mowglis_diaper
mowglis_diaper

Reputation: 509

How can I group website activity into sessions by user?

Let's say I have a dataframe like this, where there are two users and some website activity that is already grouped by session.

library(dplyr)

df <- tibble(user = c("A", "A", "A", "A", "A", "B", "B", "B", "B", "B"),
             activity = c("new", "same", "same", "new", "same", "new", "new", "same", "same", "new"))

However, I want to number the sessions by user such that they're marked like this. 

df_marked <- tibble(user = c("A", "A", "A", "A", "A", "B", "B", "B", "B", "B"),
         activity = c("new", "same", "same", "new", "same", "new", "new", "same", "same", "new"),
         session = c(1, 1, 1, 2, 2, 1, 2, 2, 2, 3))

I'm sure this is straightforward, but I don't know the relevant function in R. Thanks!

Upvotes: 0

Views: 27

Answers (1)

Sonny
Sonny

Reputation: 3183

You can use dplyr for that:

library(dplyr)
df %>%
  group_by(user) %>%
  mutate(session = cumsum(ifelse(activity == "new", 1, 0)))
# A tibble: 10 x 3
# Groups:   user [2]
   user  activity session
   <chr> <chr>      <dbl>
 1 A     new            1
 2 A     same           1
 3 A     same           1
 4 A     new            2
 5 A     same           2
 6 B     new            1
 7 B     new            2
 8 B     same           2
 9 B     same           2
10 B     new            3

Upvotes: 1

Related Questions