Jaccard

The Jaccard Coefficient Index is a measure of how similar two sets are. This library makes calculating the coefficient very easy, and provides useful helpers.

Examples

Calculate how similar two sets are:

a = ["likes:jeans", "likes:blue"]
b = ["likes:jeans", "likes:women", "likes:red"]
c = ["likes:women", "likes:red"]

# Determines how similar a pair of sets are
Jaccard.coefficient(a, b)
#=> 0.25

Jaccard.coefficient(a, c)
#=> 0.0

Jaccard.coefficient(b, c)
#=> 0.6666666666666666

# According to the input data, b and c have the most similar likes.

We can also extract the distance quite easily:

Jaccard.distance(a, b)
#=> 0.75

The Jaccard distance is the inverse relation of the coefficient: 1 - coefficient.

Find out which set is closest to a given set of attributes (return a value where the distance is the minimum):

Jaccard.closest_to(a, [b, c])
#=> ["likes:jeans", "likes:women", "likes:red"]

Jaccard.closest_to(b, [a, c])
#=> ["likes:women", "likes:red"]

Finally, we can find the best pair in a set:

require "pp"
pp Jaccard.best_match([a, b, c])
# [["likes:jeans", "likes:women", "likes:red"],
#  ["likes:women", "likes:red"]]
#=> nil

Notes on scalability

This library wasn't designed to handle millions of entries. You'll have to benchmark and see if this library meets your needs.

Note on Patches/Pull Requests

  • Fork the project.
  • Make your feature addition or bug fix.
  • Add tests for it. This is important so I don't break it in a future version unintentionally.
  • Commit, do not mess with rakefile, version, or history. (if you want to have your own version, that is fine but bump version in a commit by itself I can ignore when I pull)
  • Send me a pull request. Bonus points for topic branches.

Copyright

Copyright (c) 2010 Fran├žois Beausoleil. See LICENSE for details.