分析这四个渠道 ：Affle、Mobvista、Glispa、Batmobi
1.Affle没有任何点击行为:
head(Affle[[1]])## click_time.x activity_kind network_name
## 1 1527447336 event Affle
## 2 1527454021 event Affle
## 3 1527437658 session Adwords Display Installs
## 4 1527216072 session Affle
## 5 1527612572 reattribution Adwords Display Installs
## 6 1527361784 session Affle
## adid installed_at install_time
## 1 adf14c41-bcac-4971-a0b0-c8b295ef9b10 1527447343 1527447336
## 2 3b086480-1c53-4d4e-bf7e-233c29394eef 1527454077 1527454021
## 3 749f42ec-41c0-4f3f-b990-48c4fafa2f8f 1527157096 1527157073
## 4 4ed9d343-ba83-4856-92f2-f71f3849421a 1527216138 1527216072
## 5 03302132-27d8-4776-9178-f530d6f246a2 1527154556 1527154551
## 6 2cb187f4-c2b3-4536-9d89-7c45cc1fcaad 1527361787 1527361784这一部分数据的量有：
length(unique(Affle[[1]]$adid))## [1] 21042.Affle下载与最后一次点击间隔为0:
head(Affle[[2]] %>% filter(install_time-click_time1==0))## Warning: package 'bindrcpp' was built under R version 3.4.4## # A tibble: 6 x 17
## adid install_time click_time1 action1 network1
## <chr> <dttm> <dttm> <chr> <chr>
## 1 00014760-951a-… 2018-05-25 09:48:15 2018-05-25 09:48:15 click Affle
## 2 000f17e6-db13-… 2018-05-25 11:17:32 2018-05-25 11:17:32 click Affle
## 3 000f17e6-db13-… 2018-05-25 11:17:32 2018-05-25 11:17:32 click Affle
## 4 000f17e6-db13-… 2018-05-25 11:17:32 2018-05-25 11:17:32 click Solo
## 5 000f17e6-db13-… 2018-05-25 11:17:32 2018-05-25 11:17:32 click Solo
## 6 0015e82d-b646-… 2018-05-27 02:58:27 2018-05-27 02:58:27 click Affle
## # ... with 12 more variables: click_time2 <dttm>, action2 <chr>,
## # network2 <chr>, click_time3 <dttm>, action3 <chr>, network3 <chr>,
## # click_time4 <dttm>, action4 <chr>, network4 <chr>, click_time5 <dttm>,
## # action5 <chr>, network5 <chr>这一部分数据的量有：
length(unique((Affle[[2]] %>% filter(install_time-click_time1==0))$adid))## [1] 41073.Affle下载渠道与最后一次点击渠道不一致
head(Affle[[2]] %>% filter(network1 !='Affle'))## # A tibble: 6 x 17
## adid install_time click_time1 action1 network1
## <chr> <dttm> <dttm> <chr> <chr>
## 1 000f17e6-db13-… 2018-05-25 11:17:32 2018-05-25 11:17:32 click Solo
## 2 000f17e6-db13-… 2018-05-25 11:17:32 2018-05-25 11:17:32 click Solo
## 3 002f8ca5-0e89-… 2018-05-29 08:54:39 2018-05-29 08:54:38 click Glispa
## 4 00679b72-11bc-… 2018-05-27 10:48:31 2018-05-27 10:48:30 click Glispa
## 5 007be953-8422-… 2018-05-29 10:34:11 2018-05-29 10:33:58 click Glispa
## 6 00cb2fcf-cdda-… 2018-05-27 10:42:19 2018-05-27 10:42:17 click Glispa
## # ... with 12 more variables: click_time2 <dttm>, action2 <chr>,
## # network2 <chr>, click_time3 <dttm>, action3 <chr>, network3 <chr>,
## # click_time4 <dttm>, action4 <chr>, network4 <chr>, click_time5 <dttm>,
## # action5 <chr>, network5 <chr>这一不分的量有
length(unique((Affle[[2]] %>% filter(network1 !='Affle'))$adid))## [1] 856所以，可能有问题的占比为:
(length(unique((Affle[[2]] %>% filter(install_time-click_time1==0))$adid))+length(unique((Affle[[2]] %>% filter(install_time-click_time1==0))$adid))+length(unique((Affle[[2]] %>% filter(network1 !='Affle'))$adid)))/(length(unique(Affle[[2]]$adid)) + length(unique((Affle[[2]] %>% filter(install_time-click_time1==0))$adid)))## [1] 0.99801941.Mobvista没有任何点击行为:
head(Mobvista[[1]])## click_time.x activity_kind network_name
## 1 1528787069 session Mobvista
## 2 1528431337 session Mobvista
## 3 1527236182 event Adwords UAC Installs
## 4 1528675020 session Mobvista
## 5 1528403614 session Mobvista
## 6 1528620450 session Mobvista
## adid installed_at install_time
## 1 c7a9fbea-3bc0-4855-adf9-dcb32094ac82 1528787110 1528787069
## 2 02b4e120-1677-4130-8434-ced4086e3a23 1528431369 1528431337
## 3 5428b746-f11f-4732-87b4-d8f928f99e09 1528617610 1528617457
## 4 a8c8042e-0b88-49c8-b65e-4d2a447e0a39 1528821641 1528675020
## 5 45d160c3-2163-4083-9d10-a36dd25f5698 1528403718 1528403614
## 6 060ca671-d7a9-470c-80d6-2038271414f8 1528620490 1528620450这一部分数据的量有：
length(unique(Mobvista[[1]]$adid))## [1] 12622.Mobvista下载与最后一次点击间隔为0:
head(Mobvista[[2]] %>% filter(install_time-click_time1==0))## # A tibble: 6 x 17
## adid install_time click_time1 action1 network1
## <chr> <dttm> <dttm> <chr> <chr>
## 1 0000f255-af3b-… 2018-06-01 00:58:44 2018-06-01 00:58:44 click Mobvista
## 2 00010b6d-41f9-… 2018-06-22 00:34:50 2018-06-22 00:34:50 click Mobvista
## 3 00016e4d-4644-… 2018-06-23 00:08:05 2018-06-23 00:08:05 click Mobvista
## 4 000248c4-3177-… 2018-06-22 00:34:04 2018-06-22 00:34:04 click Mobvista
## 5 00036248-e38d-… 2018-06-21 01:18:36 2018-06-21 01:18:36 click Mobvista
## 6 00039430-7788-… 2018-06-20 01:10:03 2018-06-20 01:10:03 click Mobvista
## # ... with 12 more variables: click_time2 <dttm>, action2 <chr>,
## # network2 <chr>, click_time3 <dttm>, action3 <chr>, network3 <chr>,
## # click_time4 <dttm>, action4 <chr>, network4 <chr>, click_time5 <dttm>,
## # action5 <chr>, network5 <chr>这一部分数据的量有：
length(unique((Mobvista[[2]] %>% filter(install_time-click_time1==0))$adid))## [1] 582463.Mobvista下载渠道与最后一次点击渠道不一致
head(Mobvista[[2]] %>% filter(network1 !='Affle'))## # A tibble: 6 x 17
## adid install_time click_time1 action1 network1
## <chr> <dttm> <dttm> <chr> <chr>
## 1 0000f255-af3b-… 2018-06-01 00:58:44 2018-06-01 00:58:44 click Mobvista
## 2 00010b6d-41f9-… 2018-06-22 00:34:50 2018-06-22 00:34:50 click Mobvista
## 3 00016e4d-4644-… 2018-06-23 00:08:05 2018-06-23 00:08:05 click Mobvista
## 4 000248c4-3177-… 2018-06-22 00:34:04 2018-06-22 00:34:04 click Mobvista
## 5 00036248-e38d-… 2018-06-21 01:18:36 2018-06-21 01:18:36 click Mobvista
## 6 00039430-7788-… 2018-06-20 01:10:03 2018-06-20 01:10:03 click Mobvista
## # ... with 12 more variables: click_time2 <dttm>, action2 <chr>,
## # network2 <chr>, click_time3 <dttm>, action3 <chr>, network3 <chr>,
## # click_time4 <dttm>, action4 <chr>, network4 <chr>, click_time5 <dttm>,
## # action5 <chr>, network5 <chr>这一不分的量有
length(unique((Mobvista[[2]] %>% filter(network1 !='Mobvista'))$adid))## [1] 299所以，可能有问题的占比为:
(length(unique((Mobvista[[2]] %>% filter(install_time-click_time1==0))$adid))+length(unique((Mobvista[[2]] %>% filter(install_time-click_time1==0))$adid))+length(unique((Mobvista[[2]] %>% filter(network1 !='Mobvista'))$adid)))/(length(unique(Mobvista[[2]]$adid)) + length(unique((Mobvista[[2]] %>% filter(install_time-click_time1==0))$adid)))## [1] 0.99764241.Glispa没有任何点击行为:
head(Glispa[[1]])## click_time.x activity_kind network_name
## 1 1530094926 event Glispa
## 2 1528938534 event Glispa
## 3 1529027316 event Glispa
## 4 1529111285 event Glispa
## 5 1528938534 session Glispa
## 6 1528276402 event Glispa
## adid installed_at install_time
## 1 e7bf8d54-c60c-4713-9bc1-56f17e276ca0 1530095169 1530094926
## 2 b4a3a846-4dcd-48b9-8f25-2fed4ddc8985 1528938543 1528938534
## 3 381f643e-ff52-4707-93c8-f91ea00e1372 1529027366 1529027316
## 4 c700cf29-178d-4ba6-a415-f01f5c0dc514 1529112465 1529111285
## 5 b4a3a846-4dcd-48b9-8f25-2fed4ddc8985 1528938543 1528938534
## 6 625a6fc9-004f-4f4f-ba17-192134088874 1528277967 1528276402这一部分数据的量有：
length(unique(Glispa[[1]]$adid))## [1] 72.Glispa下载与最后一次点击间隔为0:
head(Glispa[[2]] %>% filter(install_time-click_time1==0))## # A tibble: 6 x 17
## adid install_time click_time1 action1 network1
## <chr> <dttm> <dttm> <chr> <chr>
## 1 000029d0-8b0e-… 2018-06-08 16:08:07 2018-06-08 16:08:07 click Glispa
## 2 00066fe7-34ae-… 2018-06-12 09:18:14 2018-06-12 09:18:14 click Glispa
## 3 000f4070-7174-… 2018-06-01 11:39:37 2018-06-01 11:39:37 click Glispa
## 4 0011b09d-aad5-… 2018-06-05 08:42:42 2018-06-05 08:42:42 click Glispa
## 5 001248d6-800e-… 2018-06-03 14:31:15 2018-06-03 14:31:15 click Glispa
## 6 0014620d-708a-… 2018-06-04 08:42:43 2018-06-04 08:42:43 click Glispa
## # ... with 12 more variables: click_time2 <dttm>, action2 <chr>,
## # network2 <chr>, click_time3 <dttm>, action3 <chr>, network3 <chr>,
## # click_time4 <dttm>, action4 <chr>, network4 <chr>, click_time5 <dttm>,
## # action5 <chr>, network5 <chr>这一部分数据的量有：
length(unique((Glispa[[2]] %>% filter(install_time-click_time1==0))$adid))## [1] 325343.Glispa下载渠道与最后一次点击渠道不一致
head(Glispa[[2]] %>% filter(network1 !='Affle'))## # A tibble: 6 x 17
## adid install_time click_time1 action1 network1
## <chr> <dttm> <dttm> <chr> <chr>
## 1 000029d0-8b0e-… 2018-06-08 16:08:07 2018-06-08 16:08:07 click Glispa
## 2 00066fe7-34ae-… 2018-06-12 09:18:14 2018-06-12 09:18:14 click Glispa
## 3 000f4070-7174-… 2018-06-01 11:39:37 2018-06-01 11:39:37 click Glispa
## 4 0011b09d-aad5-… 2018-06-05 08:42:42 2018-06-05 08:42:42 click Glispa
## 5 001248d6-800e-… 2018-06-03 14:31:15 2018-06-03 14:31:15 click Glispa
## 6 0014620d-708a-… 2018-06-04 08:42:43 2018-06-04 08:42:43 click Glispa
## # ... with 12 more variables: click_time2 <dttm>, action2 <chr>,
## # network2 <chr>, click_time3 <dttm>, action3 <chr>, network3 <chr>,
## # click_time4 <dttm>, action4 <chr>, network4 <chr>, click_time5 <dttm>,
## # action5 <chr>, network5 <chr>这一不分的量有
length(unique((Glispa[[2]] %>% filter(network1 !='Glispa'))$adid))## [1] 339所以，可能有问题的占比为:
(length(unique((Glispa[[2]] %>% filter(network1 !='Glispa'))$adid))+length(unique((Glispa[[2]] %>% filter(install_time-click_time1==0))$adid))+length(unique(Glispa[[1]]$adid)))/(length(unique(Glispa[[2]]$adid)) + length(unique((Glispa[[2]] %>% filter(install_time-click_time1==0))$adid)))## [1] 0.50528651.Batmobi没有任何点击行为:
head(Batmobi[[1]])## click_time.x activity_kind network_name
## 1 1529424882 session Batmobi
## 2 1529253610 event Batmobi
## 3 1529601558 session Batmobi
## 4 1528794780 event Batmobi
## 5 1529255115 session Batmobi
## 6 1529774271 event Batmobi
## adid installed_at install_time
## 1 cbb7df24-6346-46dd-8790-b74b8937884b 1529424886 1529424882
## 2 fb6a20fc-0a5f-4eea-8301-8d6e93599b53 1529294092 1529253610
## 3 254b5d00-cff4-49c9-9b4a-367769aa8edd 1529601686 1529601558
## 4 51a21df2-ed22-4452-9817-13ba01d0a09e 1528795143 1528794780
## 5 bb2236f2-ec31-441d-a0cc-1e2cace7ba76 1529255144 1529255115
## 6 33171714-6db8-4279-8b98-148027f2c673 1529774378 1529774271这一部分数据的量有：
length(unique(Batmobi[[1]]$adid))## [1] 4792.Batmobi下载与最后一次点击间隔为0:
head(Batmobi[[2]] %>% filter(install_time-click_time1==0))## # A tibble: 6 x 17
## adid install_time click_time1 action1 network1
## <chr> <dttm> <dttm> <chr> <chr>
## 1 00093036-cd9b-… 2018-06-16 00:39:45 2018-06-16 00:39:45 click Batmobi
## 2 000b2fc9-c221-… 2018-06-24 00:26:12 2018-06-24 00:26:12 click Batmobi
## 3 001169e6-9753-… 2018-06-20 01:02:23 2018-06-20 01:02:23 click Batmobi
## 4 001b6aad-a34c-… 2018-06-20 00:52:55 2018-06-20 00:52:55 click Batmobi
## 5 00229c86-abcb-… 2018-06-28 02:29:10 2018-06-28 02:29:10 click Batmobi
## 6 003526bf-f96e-… 2018-06-22 00:49:55 2018-06-22 00:49:55 click Batmobi
## # ... with 12 more variables: click_time2 <dttm>, action2 <chr>,
## # network2 <chr>, click_time3 <dttm>, action3 <chr>, network3 <chr>,
## # click_time4 <dttm>, action4 <chr>, network4 <chr>, click_time5 <dttm>,
## # action5 <chr>, network5 <chr>这一部分数据的量有：
length(unique((Batmobi[[2]] %>% filter(install_time-click_time1==0))$adid))## [1] 89353.Batmobi下载渠道与最后一次点击渠道不一致
head(Batmobi[[2]] %>% filter(network1 !='Affle'))## # A tibble: 6 x 17
## adid install_time click_time1 action1 network1
## <chr> <dttm> <dttm> <chr> <chr>
## 1 00093036-cd9b-… 2018-06-16 00:39:45 2018-06-16 00:39:45 click Batmobi
## 2 000b2fc9-c221-… 2018-06-24 00:26:12 2018-06-24 00:26:12 click Batmobi
## 3 001169e6-9753-… 2018-06-20 01:02:23 2018-06-20 01:02:23 click Batmobi
## 4 001b6aad-a34c-… 2018-06-20 00:52:55 2018-06-20 00:52:55 click Batmobi
## 5 00229c86-abcb-… 2018-06-28 02:29:10 2018-06-28 02:29:10 click Batmobi
## 6 003526bf-f96e-… 2018-06-22 00:49:55 2018-06-22 00:49:55 click Batmobi
## # ... with 12 more variables: click_time2 <dttm>, action2 <chr>,
## # network2 <chr>, click_time3 <dttm>, action3 <chr>, network3 <chr>,
## # click_time4 <dttm>, action4 <chr>, network4 <chr>, click_time5 <dttm>,
## # action5 <chr>, network5 <chr>这一不分的量有
length(unique((Batmobi[[2]] %>% filter(network1 !='Batmobi'))$adid))## [1] 1867所以，可能有问题的占比为:
(length(unique((Batmobi[[2]] %>% filter(network1 !='Batmobi'))$adid))+length(unique((Batmobi[[2]] %>% filter(install_time-click_time1==0))$adid))+length(unique(Batmobi[[1]]$adid)))/(length(unique(Batmobi[[2]]$adid)) + length(unique((Batmobi[[2]] %>% filter(install_time-click_time1==0))$adid)))## [1] 0.6209269a <- (length(unique((Batmobi[[2]] %>% filter(network1 !='Batmobi'))$adid))+length(unique((Batmobi[[2]] %>% filter(install_time-click_time1==0))$adid))+length(unique(Batmobi[[1]]$adid)))/(length(unique(Batmobi[[2]]$adid)) + length(unique((Batmobi[[2]] %>% filter(install_time-click_time1==0))$adid)))
b <- (length(unique((Glispa[[2]] %>% filter(network1 !='Glispa'))$adid))+length(unique((Glispa[[2]] %>% filter(install_time-click_time1==0))$adid))+length(unique(Glispa[[1]]$adid)))/(length(unique(Glispa[[2]]$adid)) + length(unique((Glispa[[2]] %>% filter(install_time-click_time1==0))$adid)))
c <- (length(unique((Mobvista[[2]] %>% filter(install_time-click_time1==0))$adid))+length(unique((Mobvista[[2]] %>% filter(install_time-click_time1==0))$adid))+length(unique((Mobvista[[2]] %>% filter(network1 !='Mobvista'))$adid)))/(length(unique(Mobvista[[2]]$adid)) + length(unique((Mobvista[[2]] %>% filter(install_time-click_time1==0))$adid)))
d <- (length(unique((Affle[[2]] %>% filter(install_time-click_time1==0))$adid))+length(unique((Affle[[2]] %>% filter(install_time-click_time1==0))$adid))+length(unique((Affle[[2]] %>% filter(network1 !='Affle'))$adid)))/(length(unique(Affle[[2]]$adid)) + length(unique((Affle[[2]] %>% filter(install_time-click_time1==0))$adid)))
result <- data.frame(c('Batmobi','Glispa','Mobvista','Affle'),c(a,b,c,d))
names(result) <- c('type','rate')
library(ggplot2)
library(plotly)##
## Attaching package: 'plotly'## The following object is masked from 'package:ggplot2':
##
## last_plot## The following object is masked from 'package:stats':
##
## filter## The following object is masked from 'package:graphics':
##
## layoutp <- ggplot(aes(x = type,y = rate),data = result)+geom_bar(stat= 'identity')
ggplotly(p)