R 名字的模糊匹配

发布于02月03日

在R中，我有一个第一个框架，其中有一个名为"name"的列，包含字符串，包括名字，姓氏和员工.我还有第二个名为"preusuel"的列，其中只包含名字.如何连接两个基于DF的匹配名字？

我try 的东西都没有成功.

# Install the fuzzyjoin package if not already installed
# install.packages("fuzzyjoin")

library(fuzzyjoin)
library(dplyr)

# Suppose your dataframes are named df1 and df2 with the columns "name" and "preusuel" respectively
df1 <- data.frame(name = c("Doctor John Doe", "Jane Smith", "Assistant manager Alice Johnson"))
df2 <- data.frame(preusuel = c("John", "Bob", "Alice"))

# Using fuzzy matching with stringdist_left_join
result <- stringdist_left_join(df1, df2, by = c("name" = "preusuel"), method = "jaccard", max_dist = 0.2)

我的结果是:

name	preusuel
Assistant Manager John Doe	NA
Jane Smith	NA
Doctor Johnson Alice	NA

推荐答案

像这样的东西，用grep代替？

df1 <- data.frame(name = c("Doctor John Doe", "Jane Smith", "Assistant manager Alice Johnson"))
df2 <- data.frame(preusuel = c("John", "Bob", "Alice"))

found <- sapply(df2$preusuel, \(s) {
  pat <- paste0("\\<", s, "\\>")
  i <- grep(pat, df1$name)
  if(length(i)) i else 0L
})

df1$preusuel <- NA_character_
df1$preusuel[found] <- df2$preusuel[found]
df1
#>                              name preusuel
#> 1                 Doctor John Doe     John
#> 2                      Jane Smith     <NA>
#> 3 Assistant manager Alice Johnson    Alice

^{创建于2024-02-03年第reprex v2.0.2页}