Created
June 13, 2017 10:43
-
-
Save w00lf/6ebd10d1294a9a58a40b5257e725a7f3 to your computer and use it in GitHub Desktop.
Computes cosine similarity of two text, matched by words.
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
# Inspired by: https://stackoverflow.com/questions/1746501/can-someone-give-an-example-of-cosine-similarity-in-a-very-simple-graphical-wa | |
# And: https://github.com/agarie/measurable/blob/8ff8efbab1a0892bdddf6e865dd8864956168a91/lib/measurable/cosine.rb | |
# https://github.com/agarie/measurable/blob/8ff8efbab1a0892bdddf6e865dd8864956168a91/lib/measurable/euclidean.rb | |
# Accept two text adn calcualtes cosine similarity by words of these texts, returns Float, between 0.0(not similat at all) and 1.0(identical) | |
def cosine_similarity(one, two) | |
indexes = [ one, two ].map { |text| text.scan(/[a-zA-Z]{1,}/) }.flatten.uniq | |
counters = [ one, two ].map do |text| | |
counter = text.scan(/[a-zA-Z]{1,}/).reduce(Hash.new(0)) { |x,y| x.tap {|n| n[y] += 1 } } | |
indexes.map { |key| counter[key] } | |
end | |
dot_product = counters.first.zip(counters.last).reduce(0.0) {|acc,n| acc += n[0] * n[1] } | |
# Measure euclidian product: Sqroot of (A * B) ** 2 | |
eucl_product = Math.sqrt(counters.first.reduce(0.0) {|acc,n| acc += n ** 2 }) * Math.sqrt(counters.last.reduce(0.0) {|acc,n| acc += n ** 2 }) | |
dot_product/eucl_product | |
end |
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment