Get disjoint probable duplicate sets

DisProbDup finds and joins intersecting sets in an object of class ProbDup to get disjoint probable duplicate sets.

DisProbDup(pdup, combine = c("F", "P", "S"))

Arguments

pdup

An object of class ProbDup.

combine

A character vector indicating the type of sets to be considered together for retrieving disjoint sets. If NULL, then disjoint sets within each type are retrieved (see Details).

Value

Returns an object of class ProbDup with either the disjoint sets within each type - FuzzyDuplicates, PhoneticDuplicates

and SemanticDuplicates when combine = NULL or the combined disjoint duplicate sets as an additional element DisjointDupicates

according to the choice specified in the argument combine.

Details

This function considers the accession primary keys/IDs for finding intersecting sets and subsequently joins them to retrieve disjoint sets. These operations are implemented utilizing the igraph package functions.

Disjoint sets are retrieved either individually for each type of probable duplicate sets or considering all type of sets simultaneously. In case of the latter, the disjoint of all the type of sets alone are returned in the output as an additional data frame DisjointDuplicates in an object of class ProbDup

See also

ProbDup

Examples

# \dontshow{
threads_dt <- data.table::getDTthreads()
threads_OMP <- Sys.getenv("OMP_THREAD_LIMIT")
data.table::setDTthreads(2)

data.table::setDTthreads(2)
Sys.setenv(`OMP_THREAD_LIMIT` = 2)
# }

if (FALSE) {

# Load PGR passport database
GN <- GN1000

# Specify as a vector the database fields to be used
GNfields <- c("NationalID", "CollNo", "DonorID", "OtherID1", "OtherID2")

# Clean the data
GN[GNfields] <- lapply(GN[GNfields], function(x) DataClean(x))
y1 <- list(c("Gujarat", "Dwarf"), c("Castle", "Cary"), c("Small", "Japan"),
c("Big", "Japan"), c("Mani", "Blanco"), c("Uganda", "Erect"),
c("Mota", "Company"))
y2 <- c("Dark", "Light", "Small", "Improved", "Punjab", "SAM")
y3 <- c("Local", "Bold", "Cary", "Mutant", "Runner", "Giant", "No.",
        "Bunch", "Peanut")
GN[GNfields] <- lapply(GN[GNfields], function(x) MergeKW(x, y1, delim = c("space", "dash")))
GN[GNfields] <- lapply(GN[GNfields], function(x) MergePrefix(x, y2, delim = c("space", "dash")))
GN[GNfields] <- lapply(GN[GNfields], function(x) MergeSuffix(x, y3, delim = c("space", "dash")))

# Generate KWIC index
GNKWIC <- KWIC(GN, GNfields)

# Specify the exceptions as a vector
exep <- c("A", "B", "BIG", "BOLD", "BUNCH", "C", "COMPANY", "CULTURE",
         "DARK", "E", "EARLY", "EC", "ERECT", "EXOTIC", "FLESH", "GROUNDNUT",
         "GUTHUKAI", "IMPROVED", "K", "KUTHUKADAL", "KUTHUKAI", "LARGE",
         "LIGHT", "LOCAL", "OF", "OVERO", "P", "PEANUT", "PURPLE", "R",
         "RED", "RUNNER", "S1", "SAM", "SMALL", "SPANISH", "TAN", "TYPE",
         "U", "VALENCIA", "VIRGINIA", "WHITE")

# Specify the synsets as a list
syn <- list(c("CHANDRA", "AH114"), c("TG1", "VIKRAM"))

# Fetch probable duplicate sets
GNdup <- ProbDup(kwic1 = GNKWIC, method = "a", excep = exep, fuzzy = TRUE,
                 phonetic = TRUE, encoding = "primary",
                 semantic = TRUE, syn = syn)
lapply(GNdup, dim)

# Get disjoint probable duplicate sets of each kind
disGNdup1 <- DisProbDup(GNdup, combine = NULL)
lapply(disGNdup1, nrow)

# Get disjoint probable duplicate sets combining all the kinds of sets
disGNdup2 <- DisProbDup(GNdup, combine = c("F", "P", "S"))
lapply(disGNdup2, nrow)

}

# \dontshow{
data.table::setDTthreads(threads_dt)
Sys.setenv(`OMP_THREAD_LIMIT` = threads_OMP)
# }