#!/usr/bin/python
from database import critics
# Read evaluation
print critics['Lisa Rose']['Lady in the Water']
# Set value
critics['Toby']['Snakes on a Plane'] = 5.0
print critics['Toby']
#!/usr/bin/python
from database import critics
from math import sqrt
#
# Calculate similarity between person1, person2
# using euclid distance
#
def sim_distance(prefs, person1, person2):
# Get the list item evaluated by twosome
items = [1 for item in prefs[person1] if item in prefs[person2]]
# Nothing items
if len(items) == 0: return 0
# Calculate euclid distance
distance = sum([pow(prefs[person1][item]-prefs[person2][item],2)
for item in prefs[person1] if item in prefs[person2]])
return 1.0/(1.0 + distance)
# main
if __name__ == '__main__':
print sim_distance(critics, 'Lisa Rose', 'Gene Seymour')
#!/usr/bin/python
from database import critics
from math import sqrt
#
# Calculate pearson correlation between p1 and p2
#
def sim_pearson(prefs,p1,p2):
# Get item list
items = {}
for item in prefs[p1]:
if item in prefs[p2]:
items[item]=1
n = len(items)
# return 0 when items are nothing
if n == 0: return 0
# calculate p's variance
ave1 = sum([prefs[p1][it] for it in items])/float(n)
var1 = sqrt(sum([pow(prefs[p1][it]-ave1,2) for it in items]))
ave2 = sum([prefs[p2][it] for it in items])/float(n)
var2 = sqrt(sum([pow(prefs[p2][it]-ave1,2) for it in items]))
# calculate covariance
cov = sum([(prefs[p1][it]-ave1)*(prefs[p2][it]-ave2) for it in items])
# calculate similarity
if var1*var2 == 0: return 0
return cov/(var1*var2)
if __name__ == '__main__':
print sim_pearson(critics, 'Lisa Rose', 'Gene Seymour')
#!/usr/bin/python
from database import critics
from pearson import sim_pearson
from euclid import sim_distance
#
# Calculate weighed mean, and recommend
#
def getRecommendations(prefs,person,similarity=sim_pearson):
totals = {}
simSums = {}
for other in prefs:
if other == person: continue
sim = similarity(prefs, person, other)
# ignore similarity is less than 0
if sim <= 0: continue
for item in prefs[other]:
if item not in prefs[person] or prefs[person][item] == 0:
totals.setdefault(item,0)
totals[item] += prefs[other][item] * sim
# similarity
simSums.setdefault(item,0)
simSums[item] += sim
# make normalized list
rankings = [(total/simSums[item],item) for item,total in totals.items()]
rankings.sort()
rankings.reverse()
return rankings
if __name__ == '__main__':
print getRecommendations(critics, 'Toby')
#!/usr/bin/python
from database import critics
from pearson import sim_pearson
from euclid import sim_distance
#
# Transform key and value
#
def transformPrefs(prefs):
result = {}
for person in prefs:
for item in prefs[person]:
result.setdefault(item,{})
result[item][person]=prefs[person][item]
return result
#
# Make ranking
#
def topMatches(prefs,person,n=5,similarity=sim_pearson):
scores=[(similarity(prefs,person,other),other) for other in prefs if other!=person]
scores.sort()
scores.reverse()
return scores[0:n]
if __name__ == '__main__':
movies = transformPrefs(critics)
print topMatches(movies, 'Superman Returns')
#!/usr/bin/python
from database import critics
from itemsim import transformPrefs,topMatches
from euclid import sim_distance
#
# calculate similarity between item and item
#
def calculateSimilarItems(prefs,n=10):
result = {}
itemPrefs = transformPrefs(prefs)
c = 0
for item in itemPrefs:
c += 1
if c % 100 == 0: print "%d / %d" % (c,len(itemPrefs))
scores=topMatches(itemPrefs,item,n=n,similarity=sim_distance)
result[item]=scores
return result
#
# Recommend on a item bases
#
def getRecommendedItems(prefs, itemMatch, user):
userRatings = prefs[user]
scores = {}
totalSim = {}
for (item,rating) in userRatings.items():
for (similarity,item2) in itemMatch[item]:
if item2 in userRatings: continue
scores.setdefault(item2,0)
scores[item2] += similarity*rating
totalSim.setdefault(item2,0)
totalSim[item2] += similarity
rankings = [(score/totalSim[item],item) for item,score in scores.items()]
rankings.sort()
rankings.reverse()
return rankings
if __name__ == '__main__':
itemsim = calculateSimilarItems(critics)
print getRecommendedItems(critics,itemsim,'Toby')
