python3实现基于用户的协同过滤

本文实例为大家分享了python3实现基于用户协同过滤的具体代码，供大家参考，具体内容如下

废话不多说，直接看代码。

#!/usr/bin/python3
# -*- coding: utf-8 -*-
#20170916号协同过滤电影推荐基稿
#字典等格式数据处理及直接写入文件

##from numpy import *
import time
from math import sqrt
##from texttable import Texttable

class CF:

def __init__(self, movies, ratings, k=5, n=20):
 self.movies = movies#[MovieID,Title,Genres]
 (self.train_data,self.test_data) = (ratings[0], ratings[1])#[UserID::MovieID::Rating::Timestamp]
 # 邻居个数
 self.k = k
 # 推荐个数
 self.n = n
 # 用户对电影的评分
 # 数据格式{'UserID用户ID':[(MovieID电影ID,Rating用户对电影的评星)]}
 self.userDict = {}
 # 对某电影评分的用户
 # 数据格式：{'MovieID电影ID':[UserID,用户ID]}
 # {'1',[1,2,3..],...}
 self.ItemUser = {}
 # 邻居的信息
 self.neighbors = []
 # 推荐列表
 self.recommandList = []#包含dist和电影id
 self.recommand = [] #训练集合测试集的交集，且仅有电影id
 #用户评过电影信息
 self.train_user = []
 self.test_user = []
 #给用户的推荐列表，仅含movieid
 self.train_rec =[]
 self.test_rec = []
 #test中的电影评分预测数据集合,
 self.forecast = {}#前k个近邻的评分集合
 self.score = {}#最终加权平均后的评分集合{“电影id”：预测评分}
 #召回率和准确率
 self.pre = [0.0,0.0]
 self.z = [0.0, 0.0]
'''''
userDict数据格式：
'3': [('3421', 0.8), ('1641', 0.4), ('648', 0.6), ('1394', 0.8), ('3534', 0.6), ('104', 0.8),
('2735', 0.8), ('1210', 0.8), ('1431', 0.6), ('3868', 0.6), ('1079', 1.0), ('2997', 0.6),
('1615', 1.0), ('1291', 0.8), ('1259', 1.0), ('653', 0.8), ('2167', 1.0), ('1580', 0.6),
('3619', 0.4), ('260', 1.0), ('2858', 0.8), ('3114', 0.6), ('1049', 0.8), ('1261', 0.2),
('552', 0.8), ('480', 0.8), ('1265', 0.4), ('1266', 1.0), ('733', 1.0), ('1196', 0.8),
('590', 0.8), ('2355', 1.0), ('1197', 1.0), ('1198', 1.0), ('1378', 1.0), ('593', 0.6),
('1379', 0.8), ('3552', 1.0), ('1304', 1.0), ('1270', 0.6), ('2470', 0.8), ('3168', 0.8),
('2617', 0.4), ('1961', 0.8), ('3671', 1.0), ('2006', 0.8), ('2871', 0.8), ('2115', 0.8),
('1968', 0.8), ('1136', 1.0), ('2081', 0.8)]}
ItemUser数据格式：
{'42': ['8'], '2746': ['10'], '2797': ['1'], '2987': ['5'], '1653': ['5', '8', '9'],
'194': ['5'], '3500': ['8', '10'], '3753': ['6', '7'], '1610': ['2', '5', '7'],
'1022': ['1', '10'], '1244': ['2'], '25': ['8', '9']
'''

# 将ratings转换为userDict和ItemUser
def formatRate(self,train_or_test):
 self.userDict = {}
 self.ItemUser = {}
 for i in train_or_test:#[UserID,MovieID,Rating,Timestamp]
  # 评分最高为5 除以5 进行数据归一化
##   temp = (i[1], float(i[2]) / 5)
  temp = (i[1], float(i[2]))
##   temp = (i[1], i[2])
  # 计算userDict {'用户id':[(电影id,评分),(2,5)...],'2':[...]...}一个观众对每一部电影的评分集合
  if(i[0] in self.userDict):
   self.userDict[i[0]].append(temp)
  else:
   self.userDict[i[0]] = [temp]
  # 计算ItemUser {'电影id',[用户id..],...}同一部电影的观众集合
  if(i[1] in self.ItemUser):
   self.ItemUser[i[1]].append(i[0])
  else:
   self.ItemUser[i[1]] = [i[0]]  

# 格式化userDict数据
def formatuserDict(self, userId, p):#userID为待查询目标，p为近邻对象
 user = {}
 #user数据格式为：电影id：[userID的评分，近邻用户的评分]
 for i in self.userDict[userId]:#i为userDict数据中的每个括号同81行
  user[i[0]] = [i[1], 0]
 for j in self.userDict[p]:
  if(j[0] not in user):
   user[j[0]] = [0, j[1]]#说明目标用户和近邻用户没有同时对一部电影评分
  else:
   user[j[0]][1] = j[1]#说明两者对同一部电影都有评分
 return user

# 计算余弦距离
def getCost(self, userId, p):
 # 获取用户userId和p评分电影的并集
 # {'电影ID'：[userId的评分，p的评分]} 没有评分为0
 user = self.formatuserDict(userId, p)
 x = 0.0
 y = 0.0
 z = 0.0
 for k, v in user.items():#k是键，v是值
  x += float(v[0]) * float(v[0])
  y += float(v[1]) * float(v[1])
  z += float(v[0]) * float(v[1])
 if(z == 0.0):
  return 0
 return z / sqrt(x * y)
#计算皮尔逊相似度
##  def getCost(self, userId, p):
##   # 获取用户userId和l评分电影的并集
##   # {'电影ID'：[userId的评分，l的评分]} 没有评分为0
##   user = self.formatuserDict(userId, p)
##   sumxsq = 0.0
##   sumysq = 0.0
##   sumxy = 0.0
##   sumx = 0.0
##   sumy = 0.0
##   n = len(user)
##   for k, v in user.items():
##    sumx +=float(v[0])
##    sumy +=float(v[1])
##    sumxsq += float(v[0]) * float(v[0])
##    sumysq += float(v[1]) * float(v[1])
##    sumxy += float(v[0]) * float(v[1])
##   up = sumxy -sumx*sumy/n
##   down = sqrt((sumxsq - pow(sumxsq,2)/n)*(sumysq - pow(sumysq,2)/n))
##   if(down == 0.0):
##    return 0
##   return up/down

# 找到某用户的相邻用户
def getNearestNeighbor(self, userId):
 neighbors = []
 self.neighbors = []
 # 获取userId评分的电影都有那些用户也评过分
 for i in self.userDict[userId]:#i为userDict数据中的每个括号同95行#user数据格式为：电影id：[userID的评分，近邻用户的评分]
  for j in self.ItemUser[i[0]]:#i[0]为电影编号，j为看同一部电影的每位用户
   if(j != userId and j not in neighbors):
    neighbors.append(j)
 # 计算这些用户与userId的相似度并排序
 for i in neighbors:#i为用户id
  dist = self.getCost(userId, i)
  self.neighbors.append([dist, i])
 # 排序默认是升序，reverse=True表示降序
 self.neighbors.sort(reverse=True)
 self.neighbors = self.neighbors[:self.k]#切片操作，取前k个
##  print('neighbors',len(neighbors))

# 获取推荐列表
def getrecommandList(self, userId):
 self.recommandList = []
 # 建立推荐字典
 recommandDict = {}
 for neighbor in self.neighbors:#这里的neighbor数据格式为[[dist，用户id],[],....]
  movies = self.userDict[neighbor[1]]#movies数据格式为[(电影id，评分),()，。。。。]
  for movie in movies:
   if(movie[0] in recommandDict):
    recommandDict[movie[0]] += neighbor[0]####？？？？
   else:
    recommandDict[movie[0]] = neighbor[0]

# 建立推荐列表
 for key in recommandDict:#recommandDict数据格式{电影id：累计dist，。。。}
  self.recommandList.append([recommandDict[key], key])#recommandList数据格式【【累计dist，电影id】，【】，。。。。】
 self.recommandList.sort(reverse=True)
##  print(len(self.recommandList))
 self.recommandList = self.recommandList[:self.n]
##  print(len(self.recommandList))
# 推荐的准确率
def getPrecision(self, userId):
##  print("开始！！！")
#先运算test_data，这样最终self.neighbors等保留的是后来计算train_data后的数据(不交换位置的话就得在gR函数中增加参数保留各自的neighbor)
 (self.test_user,self.test_rec) = self.getRecommand(self.test_data,userId)#测试集的用户userId所评价的电影和给该用户推荐的电影列表
 (self.train_user,self.train_rec) = self.getRecommand(self.train_data,userId)#训练集的用户userId所评价的所有电影集合(self.train_user)和给该用户推荐的电影列表(self.train_rec)
#西安电大的张海朋：基于协同过滤的电影推荐系统的构建（2015）中的准确率召回率计算
 for i in self.test_rec:
  if i in self.train_rec:
   self.recommand.append(i)
 self.pre[0] = len(self.recommand)/len(self.train_rec)
 self.z[0] = len(self.recommand)/len(self.test_rec)
 #北京交大黄宇：基于协同过滤的推荐系统设计与实现（2015）中的准、召计算
 self.recommand = []#这里没有归零的话，下面计算初始recommand不为空
 for i in self.train_rec:
  if i in self.test_user:
   self.recommand.append(i)
 self.pre[1] = len(self.recommand)/len(self.train_rec)
 self.z[1] = len(self.recommand)/len(self.test_user)
##  print(self.train_rec,self.test_rec,"20",len(self.train_rec),len(self.train_rec))
 #对同一用户分别通过训练集和测试集处理
def getRecommand(self,train_or_test,userId):
 self.formatRate(train_or_test)
 self.getNearestNeighbor(userId)
 self.getrecommandList(userId)
 user = [i[0] for i in self.userDict[userId]]#用户userId评分的所有电影集合
 recommand = [i[1] for i in self.recommandList]#推荐列表仅有电影id的集合，区别于recommandList（还含有dist）
##  print("userid该用户已通过训练集测试集处理")
 return (user,recommand)
#对test的电影进行评分预测
def foreCast(self):
 self.forecast = {}#？？？？？前面变量统一定义初始化后，函数内部是否需要该初始化？？？？
 same_movie_id = []
 neighbors_id = [i[1] for i in self.neighbors] #近邻用户数据仅含用户id的集合  

for i in self.test_user:#i为电影id，即在test里的i有被推荐到
  if i in self.train_rec:
   same_movie_id.append(i)
   for j in self.ItemUser[i]:#j为用户id,即寻找近邻用户的评分和相似度
    if j in neighbors_id:
     user = [i[0] for i in self.userDict[j]]#self.userDict[userId]数据格式：数据格式为[(电影id，评分),()，。。。。]；这里的userid应为近邻用户p
     a = self.neighbors[neighbors_id.index(j)]#找到该近邻用户的数据【dist，用户id】
     b = self.userDict[j][user.index(i)]#找到该近邻用户的数据【电影id，用户id】
     c = [a[0], b[1], a[1]]
     if (i in self.forecast):
      self.forecast[i].append(c)
     else:
      self.forecast[i] = [c]#数据格式：字典{“电影id”：【dist，评分，用户id】【】}{'589': [[0.22655856915174025, 0.6, '419'], [0.36264561173211646, 1.0, '1349']。。。}
##  print(same_movie_id)
 #每个近邻用户的评分加权平均计算得预测评分
 self.score = {}
 if same_movie_id :#在test里的电影是否有在推荐列表里，如果为空不做判断，下面的处理会报错
  for movieid in same_movie_id:
   total_d = 0
   total_down = 0
   for d in self.forecast[movieid]:#此时的d已经是最里层的列表了【】；self.forecast[movieid]的数据格式[[]]
    total_d += d[0]*d[1]
    total_down += d[0]
   self.score[movieid] = [round(total_d/total_down,3)]#加权平均后取3位小数的精度
  #在test里但是推荐没有的电影id,这里先按零计算
  for i in self.test_user:
   if i not in movieid:
    self.score[i] = [0]
 else:
  for i in self.test_user:
   self.score[i] = [0]
##  return self.score
#计算平均绝对误差MAE
def cal_Mae(self,userId):
 self.formatRate(self.test_data)
##  print(self.userDict)
 for item in self.userDict[userId]:
  if item[0] in self.score:
   self.score[item[0]].append(item[1])#self.score数据格式[[预测分，实际分]]
##  #过渡代码
##  for i in self.score:
##   pass
 return self.score
   # 基于用户的推荐
# 根据对电影的评分计算用户之间的相似度
## def recommendByUser(self, userId):
##  print("亲，请稍等片刻，系统正在快马加鞭为你运作中")   #人机交互辅助解读，
##  self.getPrecision(self,userId)

# 获取数据
def readFile(filename):
files = open(filename, "r", encoding = "utf-8")
data = []
for line in files.readlines():
 item = line.strip().split("::")
 data.append(item)
return data
files.close()
def load_dict_from_file(filepath):
_dict = {}
try:
 with open(filepath, 'r',encoding = "utf -8") as dict_file:
  for line in dict_file.readlines():
   (key, value) = line.strip().split(':')
   _dict[key] = value
except IOError as ioerr:
 print ("文件 ％s 不存在" ％ (filepath))
return _dict
def save_dict_to_file(_dict, filepath):
try:
 with open(filepath, 'w',encoding = "utf - 8") as dict_file:
  for (key,value) in _dict.items():
   dict_file.write('％s:％s\n' ％ (key, value))

except IOError as ioerr:
 print ("文件 ％s 无法创建" ％ (filepath))
def writeFile(data,filename):
with open(filename, 'w', encoding = "utf-8")as f:
 f.write(data)

# -------------------------开始-------------------------------

def start3():
start1 = time.clock()
movies = readFile("D:/d/movies.dat")
ratings = [readFile("D:/d/201709train.txt"),readFile("D:/d/201709test.txt")]
demo = CF(movies, ratings, k=20)
userId = '1000'
demo.getPrecision(userId)
## print(demo.foreCast())
demo.foreCast()
print(demo.cal_Mae(userId))
## demo.recommendByUser(ID)  #上一句只能实现固定用户查询，这句可以实现“想查哪个查哪个”，后期可以加个循环，挨个查，查到你不想查
print("处理的数据为％d条" ％ (len(ratings[0])+len(ratings[1])))
## print("____---",len(ratings[0]),len(ratings[1]))
## print("准确率： ％.2f ％％" ％ (demo.pre * 100))
## print("召回率： ％.2f ％％" ％ (demo.z * 100))
print(demo.pre)
print(demo.z)
end1 = time.clock()
print("耗费时间： ％f s" ％ (end1 - start1))
def start1():
start1 = time.clock()
movies = readFile("D:/d/movies.dat")
ratings = [readFile("D:/d/201709train.txt"),readFile("D:/d/201709test.txt")]
demo = CF(movies, ratings, k = 20)
demo.formatRate(ratings[0])
writeFile(str(demo.userDict),"D:/d/dd/userDict.txt")
writeFile(str(demo.ItemUser), "D:/d/dd/ItemUser.txt")
## save_dict_to_file(demo.userDict,"D:/d/dd/userDict.txt")
## save_dict_to_file(demo.ItemUser,"D:/d/dd/ItemUser.txt")
print("处理结束")
## with open("D:/d/dd/userDict.txt",'r',encoding = 'utf-8') as f:
##  diction = f.read()
##  i = 0
##  for j in eval(diction):
##   print(j)
##   i += 1
##   if i == 4:
##    break
def start2():
start1 = time.clock()
movies = readFile("D:/d/movies.dat")
ratings = [readFile("D:/d/201709train.txt"),readFile("D:/d/201709test.txt")]
demo = CF(movies, ratings, k = 20)
demo.formatRate_toMovie(ratings[0])
writeFile(str(demo.movieDict),"D:/d/dd/movieDict.txt")
## writeFile(str(demo.userDict),"D:/d/dd/userDict.txt")
## writeFile(str(demo.ItemUser), "D:/d/dd/ItemUser.txt")
## save_dict_to_file(demo.userDict,"D:/d/dd/userDict.txt")
## save_dict_to_file(demo.ItemUser,"D:/d/dd/ItemUser.txt")
print("处理结束")  

if __name__ == '__main__':
start1()

来源：https://blog.csdn.net/qqzhuimengren/article/details/78345891