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
from __future__ import absolute_import
from __future__ import division
from __future__ import print_function
import pymysql
import smtplib
from email.mime.text import MIMEText
from email.utils import formataddr
from email.mime.multipart import MIMEMultipart
from email.mime.application import MIMEApplication
import redis
import datetime
from pyspark import SparkConf
import time
from pyspark.sql import SparkSession
import json
import numpy as np
import pandas as pd
from pyspark.sql.functions import lit
from pyspark.sql.functions import concat_ws
from tool import *
def get_hot_search_words_tag():
try:
hot_search = """
SELECT a.keywords,
b.id,
b.tag_type
FROM api_hot_search_words a
LEFT JOIN api_tag b ON a.keywords=b.name
WHERE a.is_delete=0
AND b.tag_type+0<'4'+0
AND b.is_online=1
ORDER BY a.sorted DESC LIMIT 10
"""
mysql_results = get_data_by_mysql('172.16.30.141', 3306, 'work', 'BJQaT9VzDcuPBqkd', 'zhengxing', hot_search)
return mysql_results
except Exception as e:
print(e)
return []
def get_user_service_portrait(cl_id, all_word_tags, all_tag_tag_type, all_3tag_2tag, size=10):
try:
db_jerry_test = pymysql.connect(host='172.16.40.158', port=4000, user='root', passwd='3SYz54LS9#^9sBvC',
db='jerry_test', charset='utf8')
cur_jerry_test = db_jerry_test.cursor()
# # 用户的非搜索、支付、验证的行为
# user_df_service_sql = "select time,cl_id,score_type,tag_id,tag_referrer,action from user_new_tag_log " \
# "where cl_id ='{}' and action not in " \
# "('api/order/validate','api/settlement/alipay_callback','do_search')".format(cl_id)
# cur_jerry_test.execute(user_df_service_sql)
# 用户的非搜索行为
user_df_service_sql = "select time,cl_id,score_type,tag_id,tag_referrer,action from user_new_tag_log " \
"where cl_id ='{}' and action != 'do_search' ".format(cl_id)
cur_jerry_test.execute(user_df_service_sql)
data = list(cur_jerry_test.fetchall())
if data:
user_df_service = pd.DataFrame(data)
user_df_service.columns = ["time", "cl_id", "score_type", "tag_id", "tag_referrer", "action"]
else:
user_df_service = pd.DataFrame(columns=["time", "cl_id", "score_type", "tag_id", "tag_referrer", "action"])
# 用户的搜索行为
user_df_search_sql = "select time,cl_id,score_type,tag_id,tag_referrer,action from user_new_tag_log " \
"where cl_id ='{}' and action = 'do_search'".format(cl_id)
cur_jerry_test.execute(user_df_search_sql)
data_search = list(cur_jerry_test.fetchall())
if data_search:
user_df_search = pd.DataFrame(data_search)
user_df_search.columns = ["time", "cl_id", "score_type", "tag_id", "tag_referrer", "action"]
else:
user_df_search = pd.DataFrame(columns=["time", "cl_id", "score_type", "tag_id", "tag_referrer", "action"])
# 搜索词转成tag
# user_df_search_2_tag = pd.DataFrame(columns=list(user_df_service.columns))
for index, row in user_df_search.iterrows():
if row['tag_referrer'] in all_word_tags:
for search_tag in all_word_tags[row['tag_referrer']]:
row['tag_id'] = int(search_tag)
user_df_service = user_df_service.append(row, ignore_index=True)
break
# 增加df字段(days_diff_now, tag_type, tag2)
if not user_df_service.empty:
user_df_service["days_diff_now"] = round((int(time.time()) - user_df_service["time"].astype(float)) / (24 * 60 * 60))
user_df_service["tag_type"] = user_df_service.apply(lambda x: all_tag_tag_type.get(x["tag_id"]), axis=1)
user_df_service = user_df_service[user_df_service['tag_type'].isin(['2','3'])]
user_log_df_tag2_list = user_df_service[user_df_service['tag_type'] == '2']['tag_id'].unique().tolist()
user_df_service["tag2"] = user_df_service.apply(lambda x:
get_tag2_from_tag3(x.tag_id, all_3tag_2tag, user_log_df_tag2_list)
if x.tag_type == '3' else x.tag_id, axis=1)
user_df_service["tag2_type"] = user_df_service.apply(lambda x: all_tag_tag_type.get(x["tag2"]), axis=1)
# 算分及比例
user_df_service["tag_score"] = user_df_service.apply(
lambda x: compute_henqiang(x.days_diff_now)/get_action_tag_count(user_df_service, x.time) if x.score_type == "henqiang" else (
compute_jiaoqiang(x.days_diff_now)/get_action_tag_count(user_df_service, x.time) if x.score_type == "jiaoqiang" else (
compute_ai_scan(x.days_diff_now)/get_action_tag_count(user_df_service, x.time) if x.score_type == "ai_scan" else (
compute_ruoyixiang(x.days_diff_now)/get_action_tag_count(user_df_service, x.time) if x.score_type == "ruoyixiang" else
compute_validate(x.days_diff_now)/get_action_tag_count(user_df_service, x.time)))), axis=1)
tag_score_sum = user_df_service.groupby(by=["tag2", "tag2_type"]).agg(
{'tag_score': 'sum', 'cl_id': 'first', 'action': 'first'}).reset_index().sort_values(by=["tag_score"],
ascending=False)
tag_score_sum['weight'] = 100 * tag_score_sum['tag_score'] / tag_score_sum['tag_score'].sum()
tag_score_sum["pay_type"] = tag_score_sum.apply(
lambda x: 3 if x.action == "api/order/validate" else (
2 if x.action == "api/settlement/alipay_callback" else 1
), axis=1
)
gmkv_tag_score_sum = tag_score_sum[["tag2", "tag_score", "weight"]][:size].to_dict('record')
# 写gmkv
gm_kv_cli = redis.Redis(host="172.16.40.135", port=5379, db=2, socket_timeout=2000)
cl_id_portrait_key = "user:service_portrait_tags:cl_id:" + str(cl_id)
tag_id_list_json = json.dumps(gmkv_tag_score_sum)
gm_kv_cli.set(cl_id_portrait_key, tag_id_list_json)
gm_kv_cli.expire(cl_id_portrait_key, time=30 * 24 * 60 * 60)
# 写tidb，redis同步
stat_date = datetime.datetime.today().strftime('%Y-%m-%d')
replace_sql = """replace into user_service_portrait_tags (stat_date, cl_id, tag_list) values("{stat_date}","{cl_id}","{tag_list}")"""\
.format(stat_date=stat_date, cl_id=cl_id, tag_list=gmkv_tag_score_sum)
cur_jerry_test.execute(replace_sql)
db_jerry_test.commit()
# 写tidb 用户分层营销
# todo 不准确，因为聚合后，一个标签会有多个来源，即多个pay_type
score_result = tag_score_sum[["tag2", "cl_id", "tag_score", "weight", "pay_type"]]
score_result.rename(columns={"tag2": "tag_id", "cl_id": "device_id", "tag_score": "score"}, inplace=True)
delete_sql = "delete from api_market_personas where device_id='{}'".format(cl_id)
cur_jerry_test.execute(delete_sql)
db_jerry_test.commit()
for index, row in score_result.iterrows():
insert_sql = "insert into api_market_personas values (null, {}, '{}', {}, {}, {})".format(
row['tag_id'], row['device_id'], row['score'], row['weight'], row['pay_type'])
cur_jerry_test.execute(insert_sql)
db_jerry_test.commit()
db_jerry_test.close()
return "sucess"
except Exception as e:
print(e)
if __name__ == '__main__':
try:
db_jerry_test = pymysql.connect(host='172.16.40.170', port=4000, user='root', passwd='3SYz54LS9#^9sBvC',
db='jerry_test', charset='utf8')
cur_jerry_test = db_jerry_test.cursor()
# 获取最近30天内的用户设备id
sql_device_ids = "select distinct cl_id from user_new_tag_log " \
"where time > UNIX_TIMESTAMP(DATE_SUB(NOW(), INTERVAL 30 day))"
cur_jerry_test.execute(sql_device_ids)
device_ids_lst = [i[0] for i in cur_jerry_test.fetchall()]
db_jerry_test.close()
# 获取搜索词及其近义词对应的tag
all_word_tags = get_all_word_tags()
all_tag_tag_type = get_all_tag_tag_type()
# 3级tag对应的2级tag
all_3tag_2tag = get_all_3tag_2tag()
# 画像冷启动
hot_search_words = get_hot_search_words_tag()
hot_search_words_portrait = list()
for tag_info in hot_search_words:
tmp = dict()
tmp["tag_score"] = 10
tmp["weight"] = 10
tmp["tag2"] = tag_info["id"]
hot_search_words_portrait.append(tmp)
gm_kv_cli = redis.Redis(host="172.16.40.135", port=5379, db=2, socket_timeout=2000)
hot_search_words_portrait_portrait_key = "user:service_coldstart_tags:cl_id:"
hot_search_words_portrait_json = json.dumps(hot_search_words_portrait)
gm_kv_cli.set(hot_search_words_portrait_portrait_key, hot_search_words_portrait_json)
gm_kv_cli.expire(hot_search_words_portrait_portrait_key, time=30 * 24 * 60 * 60)
# rdd
sparkConf = SparkConf().set("spark.hive.mapred.supports.subdirectories", "true") \
.set("spark.hadoop.mapreduce.input.fileinputformat.input.dir.recursive", "true") \
.set("spark.tispark.plan.allow_index_double_read", "false") \
.set("spark.tispark.plan.allow_index_read", "true") \
.set("spark.sql.extensions", "org.apache.spark.sql.TiExtensions") \
.set("spark.tispark.pd.addresses", "172.16.40.170:2379").set("spark.io.compression.codec", "lzf") \
.set("spark.driver.maxResultSize", "8g").set("spark.sql.avro.compression.codec", "snappy")
spark = SparkSession.builder.config(conf=sparkConf).enableHiveSupport().getOrCreate()
spark.sparkContext.setLogLevel("WARN")
spark.sparkContext.addPyFile("/srv/apps/ffm-baseline_git/eda/smart_rank/tool.py")
device_ids_lst_rdd = spark.sparkContext.parallelize(device_ids_lst)
result = device_ids_lst_rdd.repartition(100).map(lambda x: get_user_service_portrait(x, all_word_tags, all_tag_tag_type, all_3tag_2tag))
result.collect()
except Exception as e:
send_email("dist_update_user_portrait_service", "dist_update_user_portrait_service", "dist_update_user_portrait_service")