class SimString::CosineMeasure

Public Instance Methods

max_feature_size(db, query_size, alpha) click to toggle source
# File lib/simstring_pure.rb, line 71
def max_feature_size(db, query_size, alpha)
  (query_size.to_f / (alpha * alpha)).floor.to_i
end
min_feature_size(db, query_size, alpha) click to toggle source
# File lib/simstring_pure.rb, line 67
def min_feature_size(db, query_size, alpha)
  (alpha * alpha * query_size).ceil.to_i
end
minimum_common_feature_count(query_size, y_size, alpha) click to toggle source
# File lib/simstring_pure.rb, line 75
def minimum_common_feature_count(query_size, y_size, alpha)
  (alpha * Math.sqrt(query_size * y_size)).ceil.to_i
end
similarity(x_feature_set, y_feature_set) click to toggle source
# File lib/simstring_pure.rb, line 79
def similarity(x_feature_set, y_feature_set)
  (x_feature_set & y_feature_set).size.to_f / Math.sqrt(x_feature_set.size * y_feature_set.size)
end