The hardware and bandwidth for this mirror is donated by METANET, the Webhosting and Full Service-Cloud Provider.
If you wish to report a bug, or if you are interested in having us mirror your free-software or open-source project, please feel free to contact us at mirror[@]metanet.ch.

comparator: Comparison Functions for Clustering and Record Linkage

comparator implements comparison functions for clustering and record linkage applications. It includes functions for comparing strings, sequences and numeric vectors. Where possible, comparators are implemented in C/C++ to ensure fast performance.

Supported comparators

String comparators:

Edit-based:

• Levenshtein(): Levenshtein distance/similarity
• DamerauLevenshtein() Damerau-Levenshtein distance/similarity
• Hamming(): Hamming distance/similarity
• OSA(): Optimal String Alignment distance/similarity
• LCS(): Longest Common Subsequence distance/similarity
• Jaro(): Jaro distance/similarity
• JaroWinkler(): Jaro-Winkler distance/similarity

Token-based:

Not yet implemented.

Hybrid token-character:

• MongeElkan(): Monge-Elkan similarity
• FuzzyTokenSet(): Fuzzy Token Set distance

Other:

• InVocabulary(): Compares strings using a reference vocabulary. Useful for comparing names.
• Lookup(): Retrieves distances/similarities from a lookup table
• BinaryComp(): Compares strings based on whether they agree/disagree exactly.

Numeric comparators:

• Euclidean(): Euclidean (L-2) distance
• Manhattan(): Manhattan (L-1) distance
• Chebyshev(): Chebyshev (L-∞) distance
• Minkowski(): Minkowski (L-p) distance

Installation

You can install the latest release from CRAN by entering:

install.packages("comparator")

The development version can be installed from GitHub using devtools:

# install.packages("devtools")
devtools::install_github("ngmarchant/comparator")

Example

A comparator is instantiated by calling its constructor function. For example, we can instantiate a Levenshtein similarity comparator that ignores differences in upper/lowercase characters as follows:

comparator <- Levenshtein(similarity = TRUE, normalize = TRUE, ignore_case = TRUE)

We can apply the comparator to character vectors element-wise as follows:

x <- c("John Doe", "Jane Doe")
y <- c("jonathon doe", "jane doe")
elementwise(comparator, x, y)
#> [1] 0.6666667 1.0000000

# shorthand for above
comparator(x, y)
#> [1] 0.6666667 1.0000000

This comparator is also defined on sequences:

x_seq <- list(c(1, 2, 1, 1), c(1, 2, 3, 4))
y_seq <- list(c(4, 3, 2, 1), c(1, 2, 3, 1))
elementwise(comparator, x_seq, y_seq)
#> [1] 0.4545455 0.7777778

# shorthand for above
comparator(x_seq, y_seq)
#> [1] 0.4545455 0.7777778

Pairwise comparisons are also supported using the following syntax:

# compare each string in x with each string in y and return a similarity matrix
pairwise(comparator, x, y, return_matrix = TRUE)
#>           [,1]      [,2]
#> [1,] 0.6666667 0.6842105
#> [2,] 0.5384615 1.0000000

# compare the strings in x pairwise and return a similarity matrix
pairwise(comparator, x, return_matrix = TRUE)
#>           [,1]      [,2]
#> [1,] 1.0000000 0.6842105
#> [2,] 0.6842105 1.0000000

These binaries (installable software) and packages are in development.
They may not be fully stable and should be used with caution. We make no claims about them.