如何高效处理字符串并与firstname、lastname数据框匹配？

现有以下三个数据集：

字符串数据集strings #

strings <- structure(list(string = c("Jennifer Rae Hancock Brown", "Lisa Smith Houston Blogger", 
"Tina Fay Las Cruces", "\t\nJamie Tucker Style Expert", "Jessica Wright Htx Satx", 
"Julie Green Lifestyle Blogger", "Mike S Thomas Football Player", 
"Tiny Fitness Houston Studio")), class = "data.frame", row.names = c(NA, 
-8L))

名字数据集firstname和lastname #

firstname <- structure(list(firstnames = c("Jennifer", "Lisa", "Tina", "Jamie", 
"Jessica", "Julie", "Mike", "George")), class = "data.frame", row.names = c(NA, 
-8L))

lastname <- structure(list(lastnames = c("Hancock", "Smith", "Houston", "Fay", 
"Tucker", "Wright", "Green", "Thomas")), class = "data.frame", row.names = c(NA, 
-8L))

需要完成以下操作：

• 1. 截取每个字符串的前三个单词，例如将"Jennifer Rae Hancock Brown"处理为"Jennifer Rae Hancock"，"Lisa Smith Houston Blogger"处理为"Lisa Smith Houston"；
• 2. 检查处理后字符串的第一个单词是否存在于firstname数据框中，匹配则生成firstname列存储匹配值，不匹配填"N/A"；
• 3. 检查处理后字符串的剩余单词是否存在于lastname数据框中，存在多个匹配则拆分多行存储对应结果。

最终要得到如下格式的final数据框：

final <- structure(list(string = c("Jennifer Rae Hancock Brown", "Lisa Smith Houston Blogger", 
"Lisa Smith Houston Blogger", "Tina Fay Las Cruces", "\t\nJamie Tucker Style Expert", 
"Jessica Wright Htx Satx", "Julie Green Lifestyle Blogger", "Mike S Thomas Football Player", 
"Tiny George Fitness Houston Studio"), firstname = c("Jennifer", 
"Lisa", "Lisa", "Tina", "Jamie", "Jessica", "Julie", "Mike", 
"N/A"), lastname = c("Hancock", "Smith", "Houston", "Fay", "Tucker", 
"Wright", "Green", "Thomas", "N/A")), class = "data.frame", row.names = c(NA, 
-9L)) 

推荐使用tidyverse工具集（dplyr+stringr+tidyr）完成，代码简洁且逻辑清晰：

步骤1：加载依赖包 #

library(tidyverse)

步骤2：预处理字符串，提取前三个单词并拆分 #

processed <- strings %>%
  # 去除字符串首尾空白（含制表符、换行符）
  mutate(clean_string = str_trim(string)) %>%
  # 提取前三个单词，兼容单词数不足3的情况
  mutate(first_three_words = str_extract(clean_string, "^\\S+\\s+\\S+\\s+\\S+|^\\S+\\s+\\S+|^\\S+")) %>%
  # 拆分第一个单词和剩余所有单词
  separate(first_three_words, into = c("first_word", "rest_words"), sep = "\\s+", extra = "merge", fill = "right") %>%
  # 将剩余单词拆分为单个单词的列表，方便后续展开多行
  mutate(rest_words = str_split(rest_words, "\\s+"))

步骤3：匹配名字并展开多行 #

final_result <- processed %>%
  # 匹配名字：判断首词是否在firstname集合中
  mutate(firstname = ifelse(first_word %in% firstname$firstnames, first_word, "N/A")) %>%
  # 将剩余单词列表展开为多行
  unnest(rest_words, keep_empty = TRUE) %>%
  # 匹配姓氏：判断单词是否在lastname集合中，无匹配则填"N/A"
  mutate(lastname = ifelse(rest_words %in% lastname$lastnames, rest_words, "N/A")) %>%
  # 处理剩余单词为空的情况，填充"N/A"
  mutate(lastname = replace_na(lastname, "N/A")) %>%
  # 保留目标格式需要的列
  select(string, firstname, lastname) %>%
  # 去重无效行
  distinct()

验证结果 #

可以用以下代码检查结果是否与目标final一致：

all.equal(final_result, final)

代码说明 #

• str_trim()：清理字符串首尾的干扰空白字符，避免影响单词提取；
• str_extract()：用正则精准匹配前三个单词，兼容单词数不足3的场景；
• separate()+str_split()：拆分首词与剩余单词，将剩余单词转为列表格式，为后续多行展开做准备；
• unnest()：快速将列表型数据展开为多行，实现多匹配结果的拆分；
• %in%：基于集合的快速匹配，效率远高于循环判断。

内容的提问来源于stack exchange，提问作者wizkids121