blockData: blockData

In fastLink: Fast Probabilistic Record Linkage with Missing Data

blockData

Description

Contains functionalities for blocking two data sets on one or more variables prior to conducting a merge.

Usage

blockData(dfA, dfB, varnames, window.block, window.size,
kmeans.block, nclusters, iter.max, n.cores)

Arguments

dfA	Dataset A - to be matched to Dataset B
dfB	Dataset B - to be matched to Dataset A
varnames	A vector of variable names to use for blocking. Must be present in both dfA and dfB
window.block	A vector of variable names indicating that the variable should be blocked using windowing blocking. Must be present in varnames.
window.size	The size of the window for window blocking. Default is 1 (observations +/- 1 on the specified variable will be blocked together).
kmeans.block	A vector of variable names indicating that the variable should be blocked using k-means blocking. Must be present in varnames.
nclusters	Number of clusters to create with k-means. Default value is the number of clusters where the average cluster size is 100,000 observations.
iter.max	Maximum number of iterations for the k-means algorithm to run. Default is 5000
n.cores	Number of cores to parallelize over. Default is NULL.

Value

A list with an entry for each block. Each list entry contains two vectors — one with the indices indicating the block members in dataset A, and another containing the indices indicating the block members in dataset B.

Examples

## Not run: 
block_out <- blockData(dfA, dfB, varnames = c("city", "birthyear"))

## End(Not run)

fastLink documentation built on Nov. 17, 2023, 9:06 a.m.