如何逐行比对数据框字符串列与列表列并统计出现频次？

你需要在tibble里逐行统计V1列的字符串在对应行V2列表中的出现次数并生成V3列，之前使用的matches <- x[!x$V1 %in% x$V2]逻辑有误：该代码是将整个V1列与所有行V2列表的合并集合做匹配，而非逐行对应检查，因此得到空数据框。

数据构造代码 #

df <- tibble::tribble(
        ~V1,                                                    ~V2,
  "oranges", c("oranges", "apples", "berries", "plums", "cherries"),
   "apples", c("oranges", "apples", "berries", "bananas", "apples"),
   "grapes", c("oranges", "apples", "berries", "plums", "cherries"),
  "berries", c("berries", "apples", "berries", "plums", "cherries"),
  )

正确实现方法 #

方法1：用dplyr逐行处理

通过rowwise()指定逐行操作，再用sum()统计匹配次数：

library(dplyr)

df <- df %>%
  rowwise() %>%
  mutate(V3 = sum(V2 == V1)) %>%
  ungroup()

方法2：用purrr批量处理

map2_int()可同时遍历V1和V2的每一组对应元素，直接返回整数结果：

library(purrr)
library(dplyr)

df <- df %>%
  mutate(V3 = map2_int(V1, V2, ~sum(.y == .x)))

运行结果 #

处理后的数据框与你给出的示例一致：

# A tibble: 4 × 3
  V1      V2                                                                 V3
  <chr>   <chr>                                                           <int>
1 oranges c("oranges", "apples", "berries", "plums", "cherries")             1
2 apples  c("oranges", "apples", "berries", "bananas", "apples")             2
3 grapes  c("oranges", "apples", "berries", "plums", "cherries")             0
4 berries c("berries", "apples", "berries", "plums", "cherries")             2

内容的提问来源于stack exchange，提问作者Munrock