-
Notifications
You must be signed in to change notification settings - Fork 0
/
Copy pathitemBasedCF_main.py
230 lines (211 loc) · 9.32 KB
/
itemBasedCF_main.py
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
import random
import math
# -*- coding:utf-8 -*-
class ItemBasedCF:
def __init__(self, datafile=None, encoding='utf-8'):
self.encoding = encoding
self.datafile = datafile
self.readData()
self.splitData(3, 47)
def readData(self, datafile=None):
"""
read the data from the data file which is a data set
"""
self.datafile = datafile or self.datafile
self.encoding = "UTF-8"
self.data = []
for line in open(self.datafile):
userid, itemid, record, mtime = line.split("\t") # 分割每行信息,导出属性
self.data.append((userid, itemid, int(record))) # 去掉时间导入到数据集data中
def splitData(self, k, seed, data=None, M=8):
"""
split the data set
testdata is a test data set
traindata is a train set
"""
self.testdata = {} # dict字典数据类型
self.traindata = {}
data = data or self.data # readDate()里的数据
random.seed(seed)
for user, item, record in self.data:
if random.randint(0, M) == k:
self.testdata.setdefault(user, {})
# testdata[user]={}
self.testdata[user][item] = record # 建立user,item二维组存储测试数据
else:
self.traindata.setdefault(user, {})
self.traindata[user][item] = record # 建立user,item二维组存储训练数据
def ItemSimilarity(self, train=None): #
"""
calculate co-rated users between items
"""
train = train or self.traindata
C = dict() # 创建空字典 https://blog.csdn.net/qq_42278791/article/details/90180538?ops_request_misc=%257B
# %2522request%255Fid%2522%253A%2522158798036719195239824207%2522%252C%2522scm%2522%253A%252220140713
# .130102334..%2522%257D&request_id=158798036719195239824207&biz_id=0&utm_source=distribute.pc_search_result
# .none-task-blog-2~all~baidu_landing_v2~default-2
N = dict()
for u, items in train.items(): # 遍历 字典项(userid,itemid)
for i in items.keys():
N.setdefault(i, 0)
N[i] += 1
for j in items.keys(): # 找出不同itemid的
if i == j:
continue
C.setdefault(i, {})
C[i].setdefault(j, 0)
C[i][j] += 1
# 单个用户的所有item关联相似度，用一次不同的item，item1与item2关联数+1(C[]),根据使用频率来判断是否相似
# N[]记录每个用户的浏览各类物品次数
self.itemSimBest = dict()
for i, related_items in C.items(): # key:C.items item1
self.itemSimBest.setdefault(i, {})
for j, cij in related_items.items(): # key:related_items c[i][j]
self.itemSimBest[i].setdefault(j, 0)
self.itemSimBest[i][j] = cij / math.sqrt(N[i] * N[j])
# item1与item2的相似度=1与2的关联数/平方根(单个用户喜欢物品总数*其余人对该物品关联)
def ItemSimilarity_IUF(self, train=None): # Inverse User Frequence
"""
calculate co-rated users between items
"""
train = train or self.traindata
C = dict()
N = dict()
for u, items in train.items():
for i in items.keys():
N.setdefault(i, 0)
N[i] += 1
for j in items.keys():
if i == j:
continue
C.setdefault(i, {})
C[i].setdefault(j, 0)
C[i][j] += 1 / math.log(1 + len(items) * 1.0) # 改进点
self.itemSimBest = dict()
for i, related_items in C.items():
self.itemSimBest.setdefault(i, {})
for j, cij in related_items.items():
self.itemSimBest[i].setdefault(j, 0)
self.itemSimBest[i][j] = cij / math.sqrt(N[i] * N[j]) # cij = C[i][j]
def recommend(self, user, train=None, k=8, nitem=40):
train = train or self.traindata
rank = dict()
ru = train.get(user) # 获取某一user中所有item_id = ru
for i, pi in ru.items():
for j, wj in sorted(self.itemSimBest[i].items(), key=lambda x: x[1], reverse=True)[0:k]:
if j in ru: # 某一user喜欢的item是否在train的数据
continue
rank.setdefault(j, 0)
rank[j] += pi * wj
return dict(sorted(rank.items(), key=lambda x: x[1], reverse=True)[0:nitem]) # 返回推荐系数从大到小的推荐物品id
def recallAndPrecision(self, train=None, test=None, k=8, nitem=10):
"""
get the recall and precision
"""
train = train or self.traindata
test = test or self.testdata
hit = 0
recall = 0
precision = 0
for user in train.keys():
tu = test.get(user, {})
rank = self.recommend(user, train=train, k=k, nitem=nitem)
for item, ratings in rank.items():
if item in tu: # 某一user喜欢的item是否在rank表中
hit += 1
recall += len(tu) # 某一user的所有item
precision += nitem
return hit / (recall * 1.0), hit / (precision * 1.0)
def coverage(self, train=None, test=None, k=8, nitem=10):
train = train or self.traindata
test = test or self.testdata
recommend_items = set()
all_items = set()
for user in train.keys():
for item in train[user].keys():
all_items.add(item) # 某一user的所有item
rank = self.recommend(user, train=train, k=k, nitem=nitem)
for item, ratings in rank.items():
recommend_items.add(item) # rank表中的item
return len(recommend_items) / (len(all_items) * 1.0)
def popularity(self, train=None, test=None, k=8, nitem=10):
"""
get the popularity
"""
train = train or self.traindata
test = test or self.testdata
item_popularity = dict()
for user, items in train.items():
for item in items.keys():
item_popularity.setdefault(item, 0)
item_popularity[item] += 1
ret = 0
n = 0
for user in train.keys():
rank = self.recommend(user, train=train, k=k, nitem=nitem)
for item, ratings in rank.items():
ret += math.log(1 + item_popularity[item])
n += 1
return ret / (n * 1.0)
def recpop(self, train=None, test=None, k=8, nitem=10):
"""
get the popularity
"""
train = train or self.traindata
test = test or self.testdata
item_popularity = dict()
for user, items in train.items():
for item in items.keys():
item_popularity.setdefault(item, 0)
item_popularity[item] += 1
ret = 0
n = 0
for user in train.keys():
rank = self.recommend(user, train=train, k=k, nitem=nitem)
for item, ratings in rank.items():
ret += math.log(1 + item_popularity[item])
n += 1
return rank, ret / (n * 1.0)
def testItemBasedCF():
cf = ItemBasedCF('C:/Users/HP/Desktop/u1.data')
cf.ItemSimilarity()
print("%.13s%3s%20s%20s%20s%20s" % (" ", 'K', "recall", 'precision', 'coverage', 'popularity'))
for k in [5, 10, 20, 40, 80, 160, 320, 640, 1280]:
recall, precision = cf.recallAndPrecision(k=k)
coverage = cf.coverage(k=k)
popularity = cf.popularity(k=k)
print("%.13s%3d%19.3f%%%19.3f%%%19.3f%%%20.3f" % (
"ItemCF ", k, recall * 100, precision * 100, coverage * 100, popularity))
def RecommandPopularity():
print("%.13s%3s%13s%16s" % (" ", "id", "popularity", "recommandlist"))
cf = ItemBasedCF('C:/Users/HP/Desktop/u1.data')
cf.ItemSimilarity_IUF()
for k in [40]:
recommandlist, popularity = cf.recpop(k=k)
# recommandlist = cf.recommend(user=id, train=None, k=k, nitem=None)
print("%.13s%13.3f" % ("ItemCF_IUF ", popularity), "%16s" % (recommandlist))
def testUserBasedCF_IUF():
cf = ItemBasedCF('C:/Users/HP/Desktop/u1.data')
cf.ItemSimilarity()
print("%.13s%3s%20s%20s%20s%20s" % (" ", 'K', "recall", 'precision', 'coverage', 'popularity'))
# for k in [10]:
# recall, precision = cf.recallAndPrecision(k=k)
# coverage = cf.coverage(k=k)
# popularity = cf.popularity(k=k)
# print("%.13s%3d%19.3f%%%19.3f%%%19.3f%%%20.3f" % (
# "ItemCF ", k, recall * 100, precision * 100, coverage * 100, popularity))
cf.ItemSimilarity_IUF()
for k in [10]:
recall, precision = cf.recallAndPrecision(k=k)
coverage = cf.coverage(k=k)
popularity = cf.popularity(k=k)
print("%.13s%3d%19.3f%%%19.3f%%%19.3f%%%20.3f" % (
"ItemCF_IUF ", k, recall * 100, precision * 100, coverage * 100, popularity))
if __name__ == "__main__":
x = int(input("Please enter an integer(1,ItemCF 2,ItemCF_IUF 3,recommand_popularity):"))
if x == 1:
testItemBasedCF()
elif x == 2:
testUserBasedCF_IUF()
elif x == 3:
RecommandPopularity()