R：基于子字符串合并两个数据帧答案

【问题标题】：R: merge two dataframes based on substringR：基于子字符串合并两个数据帧
【发布时间】：2021-02-01 11:43:07
【问题描述】：

我有两个数据框。 df1 看起来像：

           Day     Element    Incident
1   2020-04-06     3101       Check incident by SOILING
2   2020-04-02     3102       Check alarm 5662
3   2020-05-21     3101       Check energy loss by METEO ERROR
4   2020-04-02     3202       Check ACDC grid

另一个，df2，看起来像这样：

         Day     Element  Incident       Energy_loss
1 2020-04-06     3101     SOILING        0.05
2 2020-04-14     3101     SOILING        0.01
3 2020-05-21     3101     METEO ERROR    0.11
4 2020-06-15     3102     METEO ERROR    0.03

我想根据Day、Element 和Incident 列合并它们，所以我需要找到df1 中的Incident 列何时包含df2 的列Incident . df1 与df2 不匹配的行可以在Energy loss 列中留下Nan。

我已经尝试过通常的合并，但由于merge 的条件之一是子字符串，它不能正常工作。

我期望的输出是：

           Day     Element    Incident                          Energy loss
1   2020-04-06     3101       Check incident by SOILING                0.05
2   2020-04-02     3102       Check alarm 5662                          Nan
3   2020-05-21     3101       Check energy loss by METEO ERROR         0.11
4   2020-04-02     3202       Check ACDC grid                           Nan

【问题讨论】：

标签： r merge substring

【解决方案1】：

我们可以使用regex_left_join

library(dplyr)
library(fuzzyjoin)
regex_left_join(df1, df2, by = c('Day', 'Element', 'Incident')) %>% 
    select(Day = Day.x, Element = Element.x, Incident = Incident.x, Energy_loss)

-输出

#       Day Element                         Incident Energy_loss
#1 2020-04-06    3101        Check incident by SOILING        0.05
#2 2020-04-02    3102                 Check alarm 5662          NA
#3 2020-05-21    3101 Check energy loss by METEO ERROR        0.11
#4 2020-04-02    3202                  Check ACDC grid          NA

数据

df1 <- structure(list(Day = c("2020-04-06", "2020-04-02", "2020-05-21", 
"2020-04-02"), Element = c(3101L, 3102L, 3101L, 3202L), 
Incident = c("Check incident by SOILING", 
"Check alarm 5662", "Check energy loss by METEO ERROR", "Check ACDC grid"
)), class = "data.frame", row.names = c("1", "2", "3", "4"))

df2 <- structure(list(Day = c("2020-04-06", "2020-04-14", "2020-05-21", 
"2020-06-15"), Element = c(3101L, 3101L, 3101L, 3102L), Incident = c("SOILING", 
"SOILING", "METEO ERROR", "METEO ERROR"), Energy_loss = c(0.05, 
0.01, 0.11, 0.03)), class = "data.frame", row.names = c("1", 
"2", "3", "4"))

【讨论】：