def test_pearson_formula(self):
self.assertAlmostEqual(
recommendations.similarity_pearson(critics, 'Lisa Rose',
'Gene Seymour'), 0.396059017191,
5)
# The smaller the number, the closer they are.
# this example with be for Snakes and Dupree movies
distance_between_toby_lasalle = sqrt((pow(critics['Toby']['Snakes on a Plane']-critics['Toby']['You, Me and Dupree'], 2) +
(pow(critics['Mick LaSalle']['Snakes on a Plane']-critics['Mick LaSalle']['You, Me and Dupree'], 2))))
print 'distance_between_toby_lasalle: ', distance_between_toby_lasalle
# This time make the values increase when people are similar
# Returns a value between 0 and 1. 1 means they match exactly
increasing_distance_between_toby_lasalle = 1/(1 + sqrt((pow(critics['Toby']['Snakes on a Plane']-critics['Toby']['You, Me and Dupree'], 2) +
(pow(critics['Mick LaSalle']['Snakes on a Plane']-critics['Mick LaSalle']['You, Me and Dupree'], 2)))))
print 'increasing_distance_between_toby_lasalle', increasing_distance_between_toby_lasalle
print '------------------------------------------------------------------------------------------'
print 'similarity_distance', similarity_distance(critics, 'Lisa Rose', 'Gene Seymour')
print 'similarity_distance', similarity_distance(critics, 'Lisa Rose', 'Jack Matthews')
print 'similarity_distance', similarity_distance(critics, 'Lisa Rose', 'Claudia Puig')
print 'similarity_distance', similarity_distance(critics, 'Lisa Rose', 'Michael Phillips')
print 'similarity_distance', similarity_distance(critics, 'Lisa Rose', 'Mick LaSalle')
print 'similarity_distance', similarity_distance(critics, 'Lisa Rose', 'Toby')
print '------------------------------------------------------------------------------------------'
# Pearson correlation coefficient
# Measures how well two sets of data fit on the same line.
print 'similarity_pearson', similarity_pearson(critics, 'Lisa Rose', 'Gene Seymour')
print 'similarity_pearson', similarity_pearson(critics, 'Lisa Rose', 'Jack Matthews')
print 'similarity_pearson', similarity_pearson(critics, 'Lisa Rose', 'Claudia Puig')
print 'similarity_pearson', similarity_pearson(critics, 'Lisa Rose', 'Michael Phillips')
print 'similarity_pearson', similarity_pearson(critics, 'Lisa Rose', 'Mick LaSalle')
print 'similarity_pearson', similarity_pearson(critics, 'Lisa Rose', 'Toby')
def test_pearson_formula(self):
self.assertAlmostEqual(
recommendations.similarity_pearson(
critics, 'Lisa Rose', 'Gene Seymour'), 0.396059017191, 5)
