FlexRL (version 0.1.0)

launchNaive: launchNaive

Description

launchNaive

Usage

launchNaive(PIVs, encodedA, encodedB)

Value

The linkage set, a dataframe of 2 columns with indices from the first data source (A) and the second one (B) for which all the PIVs (when non-missing) matches in their values. When a PIV is missing, this method will match the record to all the records in the other file for which all other values of the PIVs match. Therefore, this 'naive' (or 'simplistic') method does not enforce the one-to-one assignment constraint of record linkage (one record in one file can at most be linked to one record in the other file). This method should only be used to judge the difficulty of the record linkage task: it gives information about the amount of duplicates between files and the discriminative power of all the PIVs together as a way to link the records.

Arguments

PIVs

A vector of size the number of Partially Identifying Variables (PIVs) with their names (as columns names in the data sources).

encodedA

One data source (encoded: the categorical values of the PIVs have to be mapped to sets of natural numbers and missing values are encoded as 0).

encodedB

The other data source (encoded).

Examples

PIVs_config = list( V1 = list(stable = TRUE),
                    V2 = list(stable = TRUE),
                    V3 = list(stable = TRUE),
                    V4 = list(stable = TRUE),
                    V5 = list( stable = FALSE,
                               conditionalHazard = FALSE,
                               pSameH.cov.A = c(),
                               pSameH.cov.B = c()) )
PIVs = names(PIVs_config)
PIVs_stable = sapply(PIVs_config, function(x) x$stable)
Nval = c(6, 7, 8, 9, 15)
NRecords = c(500, 800)
Nlinks = 300
PmistakesA = c(0.02, 0.02, 0.02, 0.02, 0.02)
PmistakesB = c(0.02, 0.02, 0.02, 0.02, 0.02)
PmissingA = c(0.007, 0.007, 0.007, 0.007, 0.007)
PmissingB = c(0.007, 0.007, 0.007, 0.007, 0.007)
moving_params = list(V1=c(),V2=c(),V3=c(),V4=c(),V5=c(0.28))
enforceEstimability = TRUE
DATA = DataCreation( PIVs_config,
                     Nval,
                     NRecords,
                     Nlinks,
                     PmistakesA,
                     PmistakesB,
                     PmissingA,
                     PmissingB,
                     moving_params,
                     enforceEstimability)
A                    = DATA$A
B                    = DATA$B
Nvalues              = DATA$Nvalues
TimeDifference       = DATA$TimeDifference
proba_same_H         = DATA$proba_same_H

TrueDelta = base::data.frame( matrix(0, nrow=0, ncol=2) )
for (i in 1:Nlinks)
{
  TrueDelta = rbind(TrueDelta, cbind(rownames(A[i,]),rownames(B[i,])))
}
true_pairs = do.call(paste, c(TrueDelta, list(sep="_")))

encodedA = A
encodedB = B

encodedA[,PIVs][ is.na(encodedA[,PIVs]) ] = 0
encodedB[,PIVs][ is.na(encodedB[,PIVs]) ] = 0

DeltaResult = launchNaive(PIVs, encodedA, encodedB)

results = base::data.frame( Results=matrix(NA, nrow=6, ncol=1) )
rownames(results) = c("tp","fp","fn","f1score","fdr","sens.")
if(nrow(DeltaResult)>1){
  linked_pairs    = do.call(paste, c(DeltaResult[,c("idxA","idxB")], list(sep="_")))
  truepositive    = length( intersect(linked_pairs, true_pairs) )
  falsepositive   = length( setdiff(linked_pairs, true_pairs) )
  falsenegative   = length( setdiff(true_pairs, linked_pairs) )
  precision       = truepositive / (truepositive + falsepositive)
  fdr             = 1 - precision
  sensitivity     = truepositive / (truepositive + falsenegative)
  f1score         = 2 * (precision * sensitivity) / (precision + sensitivity)
  results[,"Naive"] = c(truepositive,falsepositive,falsenegative,f1score,fdr,sensitivity)
}
results