Fuzzy Logic Join using two columns

Question

I'm using the r package fuzzyjoin to join two data sets. Currently I am joining on one column, and would like to join on two.

first dataset has the name of a location and a column called config
second dataset has the name of a location and two three attributes
I would like to join on two columns name and TM

I've tried adding in the column names I wish to join as a vector but this doesn't seem to work. I get an error that says:

Error: Each variable must be a 1d atomic vector or list. Problem variables: col.

#This works to join on 1 column
library(fuzzyjoin)
stringdist_inner_join(Dataset1, Data2, by ="Name", distance_col = NULL)

#Joiningontwocolumns
stringdist_inner_join(Dataset1, Dataset2, by =c("Name","TM"), distance_col = NULL)

Dataset1:

 Name           Config     TM
 ALTO D         BB         T
 CONTRA         ST         D
 EIGHT A        DD         D
 OPALAS         BB         T
 SAUSALITO Y    AA         D
 SOLANO J       ST         D

Dataset2:

 Name       Age     Rank    TM
 ALTO D     50      2       T
 ALTO D     20      6       D
 CONTRA     10      10      D
 CONTRA     15      15      T
 EIGHTH     18      21      T
 OPAL       19      4       T
 SAUSALITO  2       12      D
 SOLANO     34      43      D

try `by =c("Name"="TM")`. source: the examples in `?fuzzyjoin::stringdist_inner_join` — Adam Spannbauer, Jun 05 '17 at 15:28
Seeing the example in github, you would need to write the following: `by = c(Name = "TM")` — jazzurro, Jun 05 '17 at 15:46
@jazzurro, Thanks, it worked to a certain extent. I would say the fuzzyjoin package has issues for multiple joins. — steppermotor, Jun 06 '17 at 20:41
@steppermotor I see. I am playing with the package now. I guess I will see some issues then. Thanks for telling me this! — jazzurro, Jun 06 '17 at 23:45

score 2 · Answer 1 · answered Mar 18 '19 at 21:04

2

It took a while for me to figure out but I believe the correct syntax for multiple column joins is:

stringdist_inner_join(data1, data2, 
                      by = list(x = c("Name", "TM"), y = c("Name", "TM")), 
                      distance_col = NULL))

answered Mar 18 '19 at 21:04

Arthur Yip

5,810
2
31
50

(this is for when columns may not be named the same...) – Arthur Yip Jan 06 '21 at 07:59

Fuzzy Logic Join using two columns

1 Answers1