原版

import math
from random import randrange
# 原始数据，格式为{用户1:{电影名称1:打分1, 电影名称2:打分2,...}, 用户2:{...}}
"""
data = {
    "User1": {"film1": 5, "film2": 4, "film3": 3},
    "User2": {"film1": 4, "film2": 5, "film3": 4},
    "User3": {"film1": 4, "film2": 3, "film3": 5},
    "User4": {"film1": 2, "film2": 3, "film3": 5},
    "NewUser": {"film1": 4, "film2": 5}
}
"""
# 定义计算两个用户之间相似度的函数
def user_sim(user1, user2):
    movies = set(user1.keys()) & set(user2.keys())
    numerator = sum([user1[movie] * user2[movie] for movie in movies])
    denominator1 = sum([user1[movie] ** 2 for movie in user1.keys()])
    denominator2 = sum([user2[movie] ** 2 for movie in user2.keys()])
    denominator = math.sqrt(denominator1) * math.sqrt(denominator2)
    if not denominator:
        return 0
    else:
        return numerator / float(denominator)

# 定义为新用户进行推荐的函数
def rec_mov(data, user):
    # 找出与新用户最相似的用户
    similarities = {}
    for other_user in data:
        if other_user != user:
            similarity = user_sim(data[user], data[other_user])
            similarities[other_user] = similarity

    sorted_similarities = sorted(similarities.items(), key=lambda x: x[1], reverse=True)
    most_similar_user = sorted_similarities[0][0]
    print("最匹配的用户为:", most_similar_user)
    print()
    # 找出最相似用户看过且新用户没有看过的电影，并进行推荐
    recommendations = {}
    for movie in data[most_similar_user]:
        if movie not in data[user]:
            recommendations[movie] = data[most_similar_user][movie]

    sorted_recommendations = sorted(recommendations.items(), key=lambda x: x[1], reverse=True)
    return sorted_recommendations

# 调用函数，对NewUser进行电影推荐
data = {'User' + str(i) : {'film' + str(randrange(1,10)) : randrange(10) for j in range(randrange(15))} for i in range(10)}
NewUser = {'film' + str(randrange(1,10)) : randrange(10) for i in range(randrange(5))}
print("已有的数据为:")
for i in data:
    print(i,"=",dict(data[i].items()))
print()
data["NewUser"] = NewUser
print("需要推荐电影的用户数据为: NewUser =",NewUser)
print()
recommendations = rec_mov(data, "NewUser")
print("为用户推荐的电影为: ", end="")
for recommendation in recommendations:
    print(recommendation[0],end=" ")

精简版

import math
from random import randrange

# 定义计算两个用户之间相似度的函数
def jisuan(user1, user2):
    ms = set(user1.keys()) & set(user2.keys())
    n = sum([user1[m] * user2[m] for m in ms])
    d1 = sum([user1[m] ** 2 for m in user1.keys()])
    d2 = sum([user2[m] ** 2 for m in user2.keys()])
    d = math.sqrt(d1) * math.sqrt(d2)
    if not d:
        return 0
    else:
        return n / float(d)

# 定义为新用户进行推荐的函数
def tuijian(data, user):
    # 找出与新用户最相似的用户
    ss = {}
    for ou in data:
        if ou != user:
            s = jisuan(data[user], data[ou])
            ss[ou] = s

    sorted_ss = sorted(ss.items(), key=lambda x: x[1], reverse=True)
    mu = sorted_ss[0][0]
    print("最匹配的用户为:", mu)
    print()

    rs = {}
    for movie in data[mu]:
        if movie not in data[user]:
            rs[movie] = data[mu][movie]

    sorted_rs = sorted(rs.items(), key=lambda x: x[1], reverse=True)
    return sorted_rs

#利用随机数生成数据集
data = {'User' + str(i) : {'film' + str(randrange(1,10)) : randrange(10) for j in range(randrange(15))} for i in range(10)}
NewUser = {'film' + str(randrange(1,10)) : randrange(10) for i in range(randrange(5))}
print("已有的数据为:")
for i in data:
    print(i,"=",dict(data[i].items()))
print()
#添加新用户
data["NewUser"] = NewUser
print("需要推荐电影的用户数据为: NewUser =",NewUser)
print()

rs = tuijian(data, "NewUser")
print("为用户推荐的电影为: ", end="")
for r in rs:
    print(r[0],end=" ")