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
| #include <iostream>
#include <map>
#include <math.h>
#include <vector>
#include <time.h>
#include <string>
#include <set>
#include <algorithm>
#include <string.h>
#include <stdio.h>
using namespace std;
#define Node pair<int, int>
#define usernum 7000
#define movienum 200000
class UserBaseCF {
public:
UserBaseCF() {}
void test() {
clock_t start, stop;
start = clock();
generate_dataset("", 7);
calc_user_sim();
std::vector<pair<int, double> > v = recommend(23);
stop = clock();
printf("推荐给用户 23 的电影:\n");
printf("电影编号 推荐系数\n");
for (int i = 0; i < n_rec_movie; i++)
printf("%7d %lf\n", v[i].first, v[i].second);
printf("\n总用时: %fs\n", (double)(stop-start)/CLOCKS_PER_SEC);
}
private:
vector<Node> trainset[usernum];
vector<Node> testset[usernum];
int n_sim_user = 10;
int n_rec_movie = 10;
set<int> users;
set<int> movies;
set<int> movie2user[movienum];
map<int, double> W[usernum];
protected:
void generate_dataset(const char* filename, int k) {
FILE* fp = fopen(filename, "rt");
int userId, movieId, timestamp;
double rating;
int train_len, test_len;
train_len = test_len = 0;
srand(time(0));
while (fscanf(fp, "%d::%d::%lf::%d", &userId, &movieId, &rating, ×tamp) != EOF) {
if (rand()%10 < k) {
users.insert(userId);
movies.insert(movieId);
movie2user[movieId].insert(userId);
trainset[userId].push_back(make_pair(movieId, rating));
train_len++;
} else {
testset[userId].push_back(make_pair(movieId, rating));
test_len++;
}
}
printf("数据集分割成功...\n");
printf("训练集大小: %d, 测试集大小: %d\n", train_len, test_len);
printf("训练集电影数: %lu, 训练集用户数: %lu\n", movies.size(), users.size());
fclose(fp);
}
void calc_user_sim() {
set<int> C[usernum];
map<int, int> C_count[usernum];
int N[usernum];
memset(N, 0, sizeof N);
for (auto movie : movies) {
for (auto u : movie2user[movie]) {
N[u]++;
for (auto v : movie2user[movie]) {
if (u == v) continue;
C[u].insert(v);
C_count[u][v]++;
}
}
}
for (int u = 1; u < usernum; u++) {
for (auto related_user : C[u]) {
W[u][related_user] = C_count[u][related_user]*1.0 / sqrt(N[u]*N[related_user]);
}
}
printf("用户相似度计算完成...\n");
}
vector<pair<int, double> > recommend(int userId) {
int K = 0;
map<int, double> tmp_rank;
vector<pair<int, double> > rnk;
vector<pair<int, double> > k_sim_users;
set<int> watched_movies;
for (auto val : trainset[userId])
watched_movies.insert(val.first);
for (auto it = W[userId].begin(); it != W[userId].end(); it++) {
k_sim_users.push_back(make_pair(it->first, it->second));
}
sort(k_sim_users.begin(), k_sim_users.end(),
[](const pair<int, double> &x, const pair<int, double> &y) -> double {
return x.second > y.second;
});
for (auto it = k_sim_users.begin(); K < n_sim_user && it != k_sim_users.end(); it++) {
int similar_user = it->first;
double similarity_factor = it->second;
for (auto movie : trainset[similar_user]) {
if (watched_movies.find(movie.first) != watched_movies.end())
continue;
tmp_rank[movie.first] += similarity_factor;
}
K++;
}
for (auto it : tmp_rank) {
rnk.push_back(make_pair(it.first, it.second));
}
sort(rnk.begin(), rnk.end(),
[](const pair<int, double> &x, const pair<int, double> &y) -> double {
return x.second > y.second;
});
printf("推荐列表前 %d 项已生成...\n", n_rec_movie);
return rnk;
}
};
int main(int argc, char const *argv[])
{
UserBaseCF *test = new UserBaseCF();
test->test();
return 0;
}
|
