jaccard_probability

This is a port of the
<a href="https://docs.ropensci.org/textreuse/reference/lsh_probability.html">lsh_probability</a>
function from the
<a href="https://cran.r-project.org/package=textreuse">textreuse</a>
package, with arguments changed to reflect the hyperparameters in this
package. It gives the probability that two strings of jaccard similarity
<code>similarity</code> will be matched, given the chosen bandwidth and number of
bands.

Empowers users to fuzzily-merge data frames with millions or tens of millions of rows in minutes with low memory usage. The package uses the locality sensitive hashing algorithms developed by Datar, Immorlica, Indyk and Mirrokni (2004) <doi:10.1145/997817.997857>, and Broder (1998) <doi:10.1109/SEQUEN.1997.666900> to avoid having to compare every pair of records in each dataset, resulting in fuzzy-merges that finish in linear time.

Beniamino Green

zoomerjoin

Superlatively Fast Fuzzy Joins

Etienne Bacher

The authors of the dependency Rust crates 

jaccard_probability function

<dl><dt>similarity</dt>
<dd>the similarity of the two strings you want to compare</dd>
<dt>n_bands</dt>
<dd>The number of LSH bands used in hashing.</dd>
<dt>band_width</dt>
<dd>The number of hashes in each band.</dd></dl>

Arguments

This is a port of the
<a href='https://docs.ropensci.org/textreuse/reference/lsh_probability.html'>lsh_probability</a>
function from the
<a href='https://cran.r-project.org/package=textreuse'>textreuse</a>
package, with arguments changed to reflect the hyperparameters in this
package. It gives the probability that two strings of jaccard similarity
<code>similarity</code> will be matched, given the chosen bandwidth and number of
bands.

Find Probability of Match Based on Similarity — jaccard_probability

<dl>

<dt>similarity</dt>
<dd>the similarity of the two strings you want to compare</dd>


<dt>n_bands</dt>
<dd>The number of LSH bands used in hashing.</dd>


<dt>band_width</dt>
<dd>The number of hashes in each band.</dd>

</dl>

jaccard_probability: Find Probability of Match Based on Similarity

Description

Usage

Value

Arguments

Examples