输出结果

先看推荐结果显示

实现代码

from math import sqrt

#pearson距离

def pearson_dis(rating1, rating2):

   sum_xy = 0

   sum_x = 0

   sum_y = 0

   sum_x2 = 0

   sum_y2 = 0

   n = 0

   for key in rating1:

       if key in rating2:

           n += 1

           x = rating1[key]

           y = rating2[key]

           sum_xy += x * y

           sum_x += x

           sum_y += y

           sum_x2 += pow(x, 2)

           sum_y2 += pow(y, 2)

   # now compute denominator

   denominator = sqrt(sum_x2 - pow(sum_x, 2) / n) * sqrt(sum_y2 - pow(sum_y, 2) / n)

   if denominator == 0:

       return 0

   else:

       return (sum_xy - (sum_x * sum_y) / n) / denominator

#查找最近邻函数

def computeNearestNeighbor(username, users):

   """在给定username的情况下，计算其他用户和它的距离并排序"""

   distances = []

   for user in users:  #全用户遍历，找到两个用户，计算pearson距离，依次添加到列表内

       if user != username:  

           #distance = manhattan_dis(users[user], users[username])

           distance = pearson_dis(users[user], users[username])

           distances.append((distance, user))

   distances.sort()

   return distances

#进行推荐函数

def recommend(username, users):

   nearest = computeNearestNeighbor(username, users)[0][1]

   recommendations = []

   neighborRatings = users[nearest]

   userRatings = users[username]

   for artist in neighborRatings:

       if not artist in userRatings:

           recommendations.append((artist, neighborRatings[artist]))

   results = sorted(recommendations, key=lambda artistTuple: artistTuple[1], reverse = True)

   for result in results:

       print(result[0], result[1])

recommend('Jason', users)