r - 标记 R 中的 X 次先前出现

Question

我有一个包含许多不同社交媒体创建者 (creator_id) 的数据集。他们发布了很多次 (posting_count)，如果 ad = 1，则帖子被归类为 ad。现在我总是想将 ad = 1 之前的 3 个以前的帖子归类为 1。基本上，“goal_variable”是我想要得到的。没有循环的解决方案会很酷!!

creator_id <-c("aaa","aaa","aaa","aaa","aaa","aaa","aaa","aaa","bbb","bbb","bbb","bbb","bbb","bbb","bbb","bbb","bbb") 
posting_count <- c(143,144,145,146,147,148,149,150,90,91,92,93,94,95,96,97,98) 
ad <- c(0,0,0,0,0,0,1,0,0,0,0,0,0,0,1,0,1) 
goal_variable <- c(0,0,0,1,1,1,0,0,0,0,0,1,1,1,1,1,0)
df <- cbind(creator_id, posting_count, ad, goal_variable)

Answer 1

这是使用 map 的编程方式。基本上，对于每一行，检查当前行是否在最近的 ad == 1 之前的 3 到 1 个位置之间。

library(purrr)
library(dplyr)
df %>% 
  group_by(creator_id) %>% 
  mutate(goal_variable = map_int(row_number(), ~ any((.x - which(ad == 1)) %in% -3:-1)))

输出

# A tibble: 17 × 4
# Groups:   creator_id [2]
   creator_id posting_count    ad goal_variable
   <chr>              <dbl> <dbl>         <int>
 1 aaa                  143     0             0
 2 aaa                  144     0             0
 3 aaa                  145     0             0
 4 aaa                  146     0             1
 5 aaa                  147     0             1
 6 aaa                  148     0             1
 7 aaa                  149     1             0
 8 aaa                  150     0             0
 9 bbb                   90     0             0
10 bbb                   91     0             0
11 bbb                   92     0             0
12 bbb                   93     0             1
13 bbb                   94     0             1
14 bbb                   95     0             1
15 bbb                   96     1             1
16 bbb                   97     0             1
17 bbb                   98     1             0