authors_refine: Refines the authors code output from authors_clean()

Description Usage Arguments Examples

View source: R/authors_refine.R

Description

authors_refine This function takes the author list output after the output has been synthesized for incorrect author matches. It contains a similarity score cutoff like read_authors. This however is to further constrain the list. New values ARE NOT created, instead it filters by the sim_score column in the output file.

Usage

1
authors_refine(review, prelim, sim_score = NULL, confidence = NULL)

Arguments

review

the 'review' element from the list output by authors_clean

prelim

the 'prelim' element from the list output by authors_clean

sim_score

similarity score cut off point. Number between 0-1.

confidence

confidence score cut off point. Number between 0 - 10.

Examples

 1
 2
 3
 4
 5
 6
 7
 8
 9
10
11
12
13
14
15
16
17
18
## First gather the authors data.frame from authors_clean
data(BITR)
BITR_authors <- authors_clean(BITR)
BITR_review_df <- BITR_authors$review 
BITR_prelim_df <- BITR_authors$prelim

## If accepting the preliminary disambiguation from authors_clean without review:
refine_df <- authors_refine(BITR_review_df, BITR_prelim_df, sim_score = 0.90, confidence = 5)

## Note that 'sim_score' and 'confidence' are optional arguments and are only 
## required if changing the default values. 
refine_df <- authors_refine(BITR_review_df, BITR_prelim_df)


## If changes were made to groupID or authorID in the "_review.csv" file: 
## then incorporate those changes in a text editor, save the corrections as a new file name,
## load in to R and run `authors_refine()` with the new corrections as the review arguement.
 

embruna/refsplitr documentation built on Jan. 26, 2020, 11:01 a.m.