match_rows: Fuzzy matching of cases between linelists
In finlaycampbell/rowmatcher: Matching cases between linelists

Description Usage Arguments Value Author(s) Examples

View source: R/match_rows.R

This function matches cases between linelists on specified columns using user-specified matching thresholds.

match_rows(
  x,
  y,
  by,
  score_fun = NULL,
  rescale = TRUE,
  na_score = 0,
  output = c("scores", "merged", "review"),
  top_n = NULL,
  min_score = NULL
)

`x`	A dataframe containing the columns specified in the first column of the `by` argument.
`y`	A dataframe containing the columns specified in the second column of the `by` argument.
`by`	Linelist columns to match cases on. This can be a character vector indicating column names found in both linelists, a 2-column integer matrix indicating the pairs of columns to be matched in linelist 1 and linelist 2, or a 2-column character matrix indicating the names of the columns to be matched in linelist 1 and linelist 2.
`score_fun`	An optional list of functions for customised evaluations of matches. Each function must accept two vectors as arguments and return a numeric vector of the same length indicating the quality of the match.
`rescale`	A logical indicating whether scores for each variable should be rescaled between 0 and 1.
`na_score`	A numeric indicating the score to be assigned to NA scores. NA handling can also be specified in a variable-specific manner by providing custom scoring functions to `score_fun`.
`output`	If "scores", returns a dataframe of matched scores. If "merged", returns a merged linelist using the matched indices. If "review", returns a dataframe for manual reviewing of matches.
`top_n`	An optional integer indicating the number of matches to keep per per row of the `x` dataframe, sorted by match score.
`min_score`	An optional numeric indicating the minimum match score required to keep a match.

Depending on the value of output, a dataframe containing either the matching scores, a merged database or the matches for manual review.

Finlay Campbell (finlaycampbell93@gmail.com)

data(sample_linelists)

## examine linelists
head(sample_linelists$linelist_a)
head(sample_linelist$linelist_b)

## specify matching columns
by <- matrix(c("numeric_a", "numeric_b",
               "character_a", "character_b",
               "date_a", "date_b"),
             ncol = 2, byrow = TRUE)

## find matching case indices
matches <- match_rows(
sample_linelists$linelist_a,
sample_linelists$linelist_b,
by
)
head(matches)