#匹配物件 def transformPrefs(prefs): result={} for person in prefs: for item in prefs[person]: result.setdefault(item,{}) # result[item][person]=prefs[person][item] return result #test import recommendations import recommend_items import similarity movies=transformPrefs(recommendations.critics) print similarity.topMatches(movies,'Lady in the Water') #以影片推影评人 print recommend_items.getRecommendations(movies,'Just My Luck')
#!/usr/bin/python # Item based recommendation media system import similarity from logParser import mediaUserDict print similarity.topMatches(mediaUserDict, '904968', 10, similarity.sim_distance) print '-' * 50 print mediaUserDict print '-' * 50 matriz = similarity.getArrayFromDict(mediaUserDict) u, sigma, q = similarity.svd_components(matriz) print 'u:\n%s\n' % u print 'eigen value:\n%s\n' % sigma print 'q transposta:\n%s\n' % q print '*' * 50