https://github.com/cran/RecordLinkage
Tip revision: ec99fa9c9763e2a688afac7eeb5127bf0e94d58b authored by Andreas Borg on 11 December 2014, 00:00:00 UTC
version 0.4-6
version 0.4-6
Tip revision: ec99fa9
getExpectedSize.Rd
\name{getExpectedSize}
\alias{getExpectedSize}
\alias{getExpectedSize-methods}
\alias{getExpectedSize,RLBigDataDedup-method}
\alias{getExpectedSize,RLBigDataLinkage-method}
\alias{getExpectedSize,data.frame-method}
\title{
Estimate number of record pairs.
}
\description{
Estimates the total number of record pairs generated by a dataset and
specified blocking conditions.
}
\usage{
getExpectedSize(object, ...)
\S4method{getExpectedSize}{RLBigDataDedup}(object)
\S4method{getExpectedSize}{RLBigDataLinkage}(object)
\S4method{getExpectedSize}{data.frame}(object, blockfld = list())
}
\arguments{
\item{object}{
Either a record linkage object or a dataset.
}
\item{blockfld}{A blocking definition, such as in \code{\link{compare.dedup}}}
\item{\dots}{
Placeholder for additional arguments.
}
}
\details{
The \code{"RLBigData*"} methods are only left for backward compability.
Since version 0.4, all record pairs for such objects are generated and stored
in a disk file. The methods return the true number of record pairs.
For the \code{"data.frame"} method, estimation is based on the assumption
that agreement or disagreement of one attribute is independent of the other attributes.
\code{blockfld} is a blocking definition such as for
\code{\link{RLBigDataDedup}}.
}
\value{
The expected number of record pairs.
}
\author{
Andreas Borg, Murat Sariyar
}
\keyword{classif}