-
Notifications
You must be signed in to change notification settings - Fork 0
/
Copy pathTemproal Knowledge and Structure view.py
241 lines (213 loc) · 10.1 KB
/
Temproal Knowledge and Structure view.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
231
232
233
234
235
236
237
238
239
240
241
import math
import sqlite3
import numpy as np
import tqdm
def sort_tuples_by_second_element(tuples_array):
# 使用sorted函数和lambda表达式来排序
return sorted(tuples_array, key=lambda x: x[1])
def split(x, split_indices):
# 分割操作
parts = []
start = 0
for index in split_indices:
if index > 0:
parts.append(x[start:index])
start = index
parts.append(x[start:])
# if start < len(x):
# parts.append(x[start + 1:]) # 添加最后一个部分
return parts
def split2(x, split_indices):
x = np.array(x)
new_x = np.split(x, split_indices)
print(new_x)
return new_x
def magnitude(data_points):
# 计算向量的差
mod_list = []
x_values = [point[1] for point in data_points]
y_values = [point[0] for point in data_points]
time_diffs = (np.diff(x_values)) / 86400
score_diffs = np.diff(y_values)
for i, time_diff in enumerate(time_diffs):
# 计算向量的模
mod = math.sqrt(time_diff ** 2 + score_diffs[i] ** 2)
# for time_diff, score_diff in zip(time_diffs, score_diffs):
# # 计算向量的模
# mod = math.sqrt(time_diff ** 2 + score_diff ** 2)
mod_list.append(mod)
return mod_list
def find_Max_Min_mod(cursor):
product_set = set()
max_mod_list = []
min_mod_list = []
# min_mod = max_mod = num_iter = 0
cursor.execute('select asin from Cell_Phones_and_Accessories')
asins = cursor.fetchall()
for asin in asins:
product_set.add(asin[0])
for pro in tqdm.tqdm(product_set):
cursor.execute("select Overall, UnixReviewTime from Cell_Phones_and_Accessories where asin=?",
(pro,))
row = cursor.fetchall()
sorted_tuples = sort_tuples_by_second_element(row)
if len(sorted_tuples) > 2:
# 这里可以添加进一步处理sorted_tuples的逻辑
mod_list = magnitude(sorted_tuples)
log_mod_list_values = [math.log1p(x) if x != 0 else 0 for x in mod_list]
max_mod_list.append(max(log_mod_list_values))
min_mod_list.append(min(log_mod_list_values))
# if num_iter == len(product_set):
max_mod = max(max_mod_list)
min_mod = min(min_mod_list)
return max_mod, min_mod
def remap(file_path):
product_ids = []
user_ids = []
timestamps = []
ratings = []
fr = open(file_path + 'Burst_Dataset.txt', 'r')
fw_remap_product = open(file_path + 'remap_product.txt', 'w')
fw_remap_product.write('product_id remap_id \n')
fw_remap_reviewer = open(file_path + 'remap_reviewer.txt', 'w')
fw_remap_reviewer.write('reviewer_id remap_id \n')
fw_remap_KG = open(file_path + 'remap_KG.txt', 'w')
fw_remap_KG.write('product_id rating timestamp\n')
fw_remap_Structure = open(file_path + 'remap_Structure.txt', 'w')
fw_remap_Structure.write('remap_product_id remap_reviewer_id\n')
lines = fr.readlines()
for i in range(1, len(lines) - 1):
line = lines[i].strip('\n').split(' ')
product_ids.append(line[0])
ratings.append(line[1])
timestamps.append(line[2])
user_ids.append(line[3])
# 创建一个从ID到数字的映射字典
product_id_to_numeric = {id: i for i, id in enumerate(sorted(set(product_ids)))}
user_id_to_numeric = {id: i for i, id in enumerate(sorted(set(user_ids)))}
# 将原始ID列表转换为数字列表
numeric_product_ids = [product_id_to_numeric[id] for id in product_ids]
numeric_user_ids = [user_id_to_numeric[id] for id in user_ids]
# print(numeric_product_ids) # 输出可能是 [0, 1, 2, 3]
for remap in zip(set(user_ids), set(numeric_user_ids)):
fw_remap_reviewer.write(str(remap[0]) + ' ' + str(remap[1]) + '\n')
for remap in zip(set(product_ids), set(numeric_product_ids)):
fw_remap_product.write(str(remap[0]) + ' ' + str(remap[1]) + '\n')
for remap in zip(numeric_product_ids, ratings, timestamps):
fw_remap_KG.write(str(remap[0]) + ' ' + str(remap[1]) + ' ' + str(remap[2]) + '\n')
for remap in zip(numeric_product_ids, numeric_user_ids):
fw_remap_Structure.write(str(remap[0]) + ' ' + str(remap[1]) + '\n')
fr.close()
fw_remap_product.close()
fw_remap_reviewer.close()
fw_remap_KG.close()
fw_remap_Structure.close()
class struct_kg_views:
def __init__(self, database_path, time_session_threshold):
self.database_path = database_path
self.normal_mod_list_all = set()
self.time_session_threshold = time_session_threshold
# 使用连接池来避免每次查询都重新连接数据库
self.conn = sqlite3.connect(self.database_path, check_same_thread=False)
self.cursor = self.conn.cursor()
self.file_path = 'Data/Cell_Phones_and_Accessories/All\\'
self.fw = open(self.file_path + 'Burst_Dataset.txt', 'w')
self.fw.write('Product_id' + ' ' + 'Rating' + ' ' + 'Timestamp' + ' ' + 'Reviewer_id' + '\n')
self.fw_KG = open(self.file_path + 'KG.txt', 'w')
self.fw_KG.write('Product_id' + ' ' + 'Rating' + ' ' + 'Timestamp' + '\n')
self.fw_Struct = open(self.file_path + 'Struct.txt', 'w')
self.fw_Struct.write('Product_id' + 'Reviewer_id' + '\n')
self.fw_Struct_KG_view = open(self.file_path + 'Struct_KG.txt', 'w')
self.fw_Struct_KG_view.write(
'{product:[reviewer1, reviewer2, Rating1, Rating2, timeStamp1, timestamp2, vector mod for rating and '
'timestamp of difference]}' + '\n')
def list_split(self, pro, sorted_tuples, split_indices):
# sorted_tuples is the Overall, UnixReviewTime, reviewerID sort by timeStamp
# pro is the product of the sort_tuples
# split_indices is split indexs list for sort_tuples
sessions = split(sorted_tuples, split_indices)
# print("分割后session：", sessions)
# print("分割后session：len", len(sessions))
for j, session in enumerate(sessions):
# print(f"第{j}个分割后：", session)
if len(session) > 2:
for w in range(0, len(session) - 1):
self.fw.write(str(pro) + ' ' + str(session[w][0]) + ' ' + str(session[w][1]) + ' ' + str(
session[w][2]) + '\n')
self.fw.flush()
self.fw_KG.write(str(pro) + ' ' + str(session[w][0]) + ' ' + str(session[w][1]) + '\n')
self.fw_KG.flush()
self.fw_Struct.write(str(pro) + ' ' + str(session[w][1]) + '\n')
self.fw_Struct.flush()
def Struct_KG_view(self):
max_mod = 5.86
min_mod = 0
product_set = set()
self.cursor.execute('select asin from Cell_Phones_and_Accessories_2010')
asins = self.cursor.fetchall()
for asin in asins:
product_set.add(asin[0])
# print(product_set)
print("normal its...")
# for pro in tqdm.tqdm(product_set):
for pro in product_set:
product_reviewer_magnitude = {}
normal_mod_list = []
split_indices = []
self.cursor.execute(
"select Overall, UnixReviewTime, reviewerID from Cell_Phones_and_Accessories_2010 where asin=?",
(pro,))
row = self.cursor.fetchall()
sorted_tuples = sort_tuples_by_second_element(row)
if len(sorted_tuples) > 2:
# print(sorted_tuples)
# 这里可以添加进一步处理sorted_tuples的逻辑
mod_list = magnitude(sorted_tuples)
log_mod_list_values = [math.log1p(x) if x != 0 else 0 for x in mod_list]
for i, mod in enumerate(log_mod_list_values):
mod = (mod - min_mod) / (max_mod - min_mod)
normal_mod_list.append(mod)
self.normal_mod_list_all.add(mod)
if mod > self.time_session_threshold:
split_indices.append(i + 1)
# print("归一化列表: ", normal_mod_list)
# print("len normal_mod_list", len(normal_mod_list))
# print("split_indices", split_indices)
# print("len split_indices", len(split_indices))
# print("分割前session：", sorted_tuples)
# print("分割前session：len ", len(sorted_tuples))
product_reviewer_magnitude[pro] = []
i = 0
while i < len(sorted_tuples) - 1:
product_reviewer_magnitude[pro].append(
(sorted_tuples[i][2], sorted_tuples[i + 1][2], sorted_tuples[i][0]
, sorted_tuples[i + 1][0], sorted_tuples[i][1], sorted_tuples[i + 1][1]))
# product_reviewer_magnitude format as -{'product':[reviewer1,reviewer2,Rating1,Rating2,timeStamp1,
# timestamp2,vector mod for rating and timestamp of difference]}
i += 1
for i, reviewer_tuple in enumerate(product_reviewer_magnitude[pro]):
normal_mod_list[i] = "{:.4f}".format(normal_mod_list[i])
product_reviewer_magnitude[pro][i] = (*reviewer_tuple, normal_mod_list[i])
# print(product_reviewer_magnitude)
self.fw_Struct_KG_view.write(str(product_reviewer_magnitude) + '\n')
self.list_split(pro, sorted_tuples, split_indices)
return self.normal_mod_list_all
# sessions.clear()
# self.split_array(normal_integral_area)
# print(normal_differences)
# print('评分差序列', difference)
def close(self):
self.fw.close()
self.fw_KG.close()
self.fw_Struct.close()
self.fw_Struct_KG_view.close()
# 确保在结束时关闭连接
self.conn.close()
if __name__ == '__main__':
test = struct_kg_views('G:\\DB\\amazonReviews.db', 0.0)
print("differences the asin sequences...")
normal_mod_list_all = test.Struct_KG_view()
print(len(normal_mod_list_all))
test.close()
remap(test.file_path)
# test.close() # 使用完毕后关闭数据库连接