blockMetrics: blockMetrics

In kailin-lu/recordlinkR: recordlinkR

Description

Reports block metrics to assess quality of blocking. If known matches are null then

Usage

blockMetrics(dfA, dfB, blocks, known.matches = NULL)

Arguments

dfA	dataframe A
dfB	dataframe B
blocks	Dataframe of two columns representing pairs to be checked post blocking
known.matches	Dataframe of two columns representing indices of known matches where the left index refers to a record index in dfA and the right index refers to a record index in dfB

Value

List of metrics

original.comparisons	Number of original comparisons
remaining.comparisons	Number of comparisons remaining post blocking
pairs.reduction	Proportion of original comparisons remaining post blocking
total.known.matches	If known.matches was provided, number of total known matches
found.matches	Number of known matches found in the remaning pairs
pairs.completeness	Percentage of known matches found in remaining pairs

kailin-lu/recordlinkR documentation built on May 4, 2019, 7:37 a.m.