每周从 user_id 和 date 查找新的活跃用户

Posted 2023-03-11

【中文标题】每周从 user_id 和 date 查找新的活跃用户

【英文标题】：Find new and active user each week from user_id and date

【发布时间】：2019-09-19 21:27:25

【问题描述】：

背景

现在假设我们经营基于订阅的业务（我们确实这样做了）。当客户订阅我们的产品时，他们有许多定制选项。 出于本练习的目的，我们将假设以下内容：

● 当用户注册时，将在该用户的订单表中生成一条记录。

○ 这将是该 user_id 在 Orders 中的第一条记录

○ Orders 表中的第一个日期将是用户注册的日期。

● 用户的第一个订单在他们注册的同一天发货

● 用户可以随时更改配送频率，甚至可以要求配送额外的箱子。

○ 对于这个任务，我们不关心交付的频率；主要是因为这个例子中的数据是随机生成的，在这个数据集中观察到的频率节奏违背了自然逻辑；）

● 如果或当用户取消时，他们会在订单表中的最后一个订单后的 14 天（含）内保持“活跃”状态。 ● 从第一个订单到最后一个订单之间的所有日子里，用户都被认为是“活跃的”。

○ 对于这个任务，我们不会关心探索“重新激活”；即，与已取消然后在未来某个日期重新注册的用户一起使用。为简单起见，我们将这些用户视为从未取消过。

定义

● 将用户群组定义为在同一时期首次活跃的用户的集合。

● 将给定同类群组在一段时间内的留存率定义为比率：N / D，其中 N = 同类群组中本时段活跃且上一时段也活跃的用户数 D = 同类群组中的用户数上一期活跃的人

● 根据问题中指定的内容，将期间定义为从星期日开始的日历月或日历周。

问题

生成包含列的表：

日期 |新计数 | count_active

count_new：每周有多少新用户注册？

count_active：每周有多少活跃用户？

部分数据：

    id user_id total       date payment_status
1       1       1 12783 2017-01-01           paid
2     258       1 12783 2017-01-22           paid
3    1072       1 12783 2017-02-26           paid
4    2086       1 12783 2017-03-26           paid
5    2387       1 12783 2017-04-02           paid
6    3860       1 12783 2017-04-30           paid
7    5546       1 12783 2017-05-28           paid
8       2       2  9516 2017-01-01           paid
9      68       2  9516 2017-01-08           paid
10      3       3 14536 2017-01-01           paid
11    372       3 14536 2017-01-29           paid
12    879       3 14536 2017-02-19           paid
13   1796       3 14536 2017-03-19           paid
14   3451       3 14536 2017-04-23           paid
15   4651       3 14536 2017-05-14           paid
16   5547       3 14536 2017-05-28           paid
17   6920       3 14536 2017-06-18           paid
18   7385       3 14536 2017-06-25           paid
19  10024       3 14536 2017-07-30         unpaid
20  11581       3 14536 2017-07-30         unpaid
21  13138       3 14536 2017-07-30         unpaid
22  14695       3 14536 2017-07-30         unpaid
23      4       4  5755 2017-01-01           paid
24    497       4  5755 2017-02-05           paid
25   1285       4  5755 2017-03-05           paid
26   2699       4  5755 2017-04-09           paid
27   3057       4  5755 2017-04-16           paid
28      5       5 10102 2017-01-01           paid
29    498       5 10102 2017-02-05           paid
30   1529       5 10102 2017-03-12           paid
31   2087       5 10102 2017-03-26           paid
32   2388       5 10102 2017-04-02           paid
33      6       6 13552 2017-01-01           paid
34     69       6 13552 2017-01-08           paid




structure(list(id = 1:100, user_id = c(1L, 2L, 3L, 4L, 5L, 6L, 
7L, 8L, 9L, 10L, 11L, 12L, 13L, 14L, 15L, 16L, 17L, 18L, 19L, 
20L, 21L, 22L, 23L, 24L, 25L, 26L, 27L, 28L, 29L, 30L, 31L, 32L, 
33L, 34L, 35L, 36L, 37L, 38L, 39L, 40L, 41L, 42L, 43L, 44L, 45L, 
46L, 47L, 48L, 49L, 50L, 51L, 52L, 53L, 54L, 55L, 56L, 57L, 58L, 
59L, 60L, 61L, 62L, 63L, 64L, 65L, 66L, 67L, 2L, 6L, 10L, 12L, 
17L, 21L, 27L, 29L, 36L, 37L, 40L, 49L, 55L, 59L, 61L, 67L, 68L, 
69L, 70L, 71L, 72L, 73L, 74L, 75L, 76L, 77L, 78L, 79L, 80L, 81L, 
82L, 83L, 84L), total = c(12783L, 9516L, 14536L, 5755L, 10102L, 
13552L, 6940L, 12154L, 14639L, 8034L, 10912L, 12255L, 8016L, 
6483L, 9841L, 14813L, 10934L, 5194L, 7753L, 5544L, 13813L, 9739L, 
13630L, 5281L, 10607L, 14873L, 13441L, 12998L, 10162L, 8110L, 
8269L, 9118L, 12308L, 14144L, 5789L, 7364L, 11921L, 5276L, 11695L, 
6669L, 7872L, 12890L, 7636L, 11682L, 14620L, 10876L, 12273L, 
14560L, 6787L, 13150L, 5559L, 13086L, 6957L, 6862L, 12442L, 10948L, 
12293L, 8398L, 8796L, 14986L, 6235L, 12077L, 5013L, 11953L, 7891L, 
13551L, 14988L, 9516L, 13552L, 8034L, 12255L, 10934L, 13813L, 
13441L, 10162L, 7364L, 11921L, 6669L, 6787L, 12442L, 8796L, 6235L, 
14988L, 10769L, 10875L, 10603L, 12522L, 5475L, 9343L, 6860L, 
11969L, 7392L, 9487L, 13016L, 6284L, 9801L, 6581L, 9164L, 11898L, 
9210L), date = structure(c(17167, 17167, 17167, 17167, 17167, 
17167, 17167, 17167, 17167, 17167, 17167, 17167, 17167, 17167, 
17167, 17167, 17167, 17167, 17167, 17167, 17167, 17167, 17167, 
17167, 17167, 17167, 17167, 17167, 17167, 17167, 17167, 17167, 
17167, 17167, 17167, 17167, 17167, 17167, 17167, 17167, 17167, 
17167, 17167, 17167, 17167, 17167, 17167, 17167, 17167, 17167, 
17167, 17167, 17167, 17167, 17167, 17167, 17167, 17167, 17167, 
17167, 17167, 17167, 17167, 17167, 17167, 17167, 17167, 17174, 
17174, 17174, 17174, 17174, 17174, 17174, 17174, 17174, 17174, 
17174, 17174, 17174, 17174, 17174, 17174, 17174, 17174, 17174, 
17174, 17174, 17174, 17174, 17174, 17174, 17174, 17174, 17174, 
17174, 17174, 17174, 17174, 17174), class = "Date"), payment_status = c("paid", 
"paid", "paid", "paid", "paid", "paid", "paid", "paid", "paid", 
"paid", "paid", "paid", "paid", "paid", "paid", "paid", "paid", 
"paid", "paid", "paid", "paid", "paid", "paid", "paid", "paid", 
"paid", "paid", "paid", "paid", "paid", "paid", "paid", "paid", 
"paid", "paid", "paid", "paid", "paid", "paid", "paid", "paid", 
"paid", "paid", "paid", "paid", "paid", "paid", "paid", "paid", 
"paid", "paid", "paid", "paid", "paid", "paid", "paid", "paid", 
"paid", "paid", "paid", "paid", "paid", "paid", "paid", "paid", 
"paid", "paid", "paid", "paid", "paid", "paid", "paid", "paid", 
"paid", "paid", "paid", "paid", "paid", "paid", "paid", "paid", 
"paid", "paid", "paid", "paid", "paid", "paid", "paid", "paid", 
"paid", "paid", "paid", "paid", "paid", "paid", "paid", "paid", 
"paid", "paid", "paid")), row.names = c(NA, 100L), class = "data.frame")

【问题讨论】：

欢迎来到 SO。您需要分享一个可重现的示例。使用dput分享数据

我已经添加了

【参考方案1】：

所以我设法通过检查 user_id 的第一次出现来计算 count_new 然后与初始数据合并，添加一个列，按日期和 id 告诉用户是否是新用户，然后我按日期计算新用户：

library(dplyr)
firstshow<-Orders %>%
group_by(user_id) %>%
  arrange(date) %>%
  slice(1L) %>%
  mutate(new = "new")

newdata<-merge.data.frame(Orders,firstshow,by=c("date","user_id"),all = T)
count<-newdata %>%
  filter(new=="new" ) %>%
  group_by(date) %>%
 tally()
names(count)[2]<-"count_new"