根据 R 中不同列的值的条件进行计算的函数

Question

我需要帮助来弄清楚如何在 R 中执行此操作。

我有一个 df 看起来像：

ID	ZIP code	loc 1 zip code	loc 2 zip code	loc 3 zip code	Opinion on loc 1	Opinion on loc 2	Opinion on loc 3
1	90804	91723	90814	91604	Neutral	Like	Dislike

我正在尝试添加如下所示的三列：

ID	ZIP code	loc 1 zip code	loc 2 zip code	loc 3 zip code	Opinion on loc 1	Opinion on loc 2	Opinion on loc 3	Distance from Like	Distance from Disike	Distance from Neutral
1	90804	91723	90814	91604	Neutral	Like	Dislike	42 Km	56 Km	24 Km

需要创建的列需要先匹配客户'Like'在哪个位置，计算他们与它的距离，然后将距离放入'Distance from Like'列等等。

我正在使用这个函数来计算距离：

library(zipcodeR)
zip_distance(zipcode_a, zipcode_b, lonlat = TRUE, units = "meters")

（来源：Conditional column value calculation, zip code to lat/long 上的第一个答案）

Answer 1

如果您更改您的列名，使它们的格式与 location_variable 一致，那么您可以使用 tidyr 的 pivot_long 将所有内容变成长格式，然后您可以计算所有距离作为所有位置的邮政编码现在都在一列中。然后，您可以将其保留为长数据格式，也可以将其恢复为您请求的宽格式

library(tidyverse)
test<- read.table( text = "ID,UserZIPcode,loc1_zipcode,loc2_zipcode,loc3_zipcode,loc1_Opinion,loc2_Opinion,loc3_Opinion
1,1,90804,91723,90814,91604,Neutral,Like,Dislike", sep = "," )

test_distances <- test %>% 
        pivot_longer(cols = c(-ID, -UserZIPcode), names_pattern  = "(.*)_(.*)", names_to = c("location",".value")) %>% 
        mutate(Distance = zip_distance(UserZIPcode, zipcode, lonlat = TRUE, units = "meters"))
  


test_distances %>% 
  pivot_wider(id_cols = ID, names_from = Opinion, names_prefix = "Distance_from", values_from = "Distance") %>% 
  right_join(test)

根据 R 中不同列的值的条件进行计算的函数

Function for calculating based on condition from value from different columns in R

r

dplyr

tidyr