Script 349: Mediatype tagging

Purpose

Python script for tagging media types and subtypes in a dataset.

To Elaborate

This Python script is designed to tag media types and subtypes in a dataset based on specific business rules. The script takes an input dataset and assigns media types and subtypes to each row based on the values in certain columns. The tagged dataset is then outputted for further analysis or processing.

Walking Through the Code

  1. The script starts by defining column constants and client timezone.
  2. It then creates a temporary media type and subtype variable.
  3. The input dataset is modified to include the temporary media type and subtype columns.
  4. The current date is obtained and stored in the “today” variable.
  5. The input dataset is printed using the “tableize” function.
  6. The script uses conditional statements to assign media types and subtypes based on specific business rules. Some examples include:
    • If the publisher name contains “Taboola”, the media type and subtype are set to “Native”.
    • If the publisher name contains “TikTok”, the media type and subtype are set to “Social” and “TikTok” respectively.
    • If the publisher name contains “Facebook”, the media type and subtype are set to “Social” and “Meta” respectively.
    • Similar conditional statements are used for other media types and subtypes.
  7. The modified input dataset is printed again.
  8. The script checks for duplicated index values in the input dataset.
  9. The media type and subtype columns in the output dataset are updated with the values from the temporary columns in the input dataset.
  10. The output dataset is filtered to remove rows with null values in the media type and subtype columns, as well as rows containing certain characters in the campaign column.
  11. The final output dataset is printed using the “tableize” function.

Vitals

  • Script ID : 349
  • Client ID / Customer ID: 1306925585 / 60269545
  • Action Type: Bulk Upload
  • Item Changed: Campaign
  • Output Columns: Account, Campaign, Media sub-type, Media Type
  • Linked Datasource: M1 Report
  • Reference Datasource: None
  • Owner: Tom McCaughey (tmccaughey@marinsoftware.com)
  • Created by Tom McCaughey on 2023-10-12 16:16
  • Last Updated by Tom McCaughey on 2024-04-18 10:50
> See it in Action

Python Code

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

RPT_COL_ACCOUNT = 'Account'
RPT_COL_CAMPAIGN = 'Campaign'
RPT_COL_GROUP = 'Group'
RPT_COL_MEDIA_SUBTYPE = 'Media sub-type'
RPT_COL_MEDIA_TYPE = 'Media Type'
RPT_COL_ACCOUNT_PUBLISHER_NAME = 'Publisher Name'
RPT_COL_ACCOUNT_PUBLISHER = 'Publisher'
BULK_COL_ACCOUNT = 'Account'
BULK_COL_CAMPAIGN = 'Campaign'
BULK_COL_MEDIA_SUBTYPE = 'Media sub-type'
BULK_COL_MEDIA_TYPE = 'Media Type'
BULK_COL_ACCOUNT_PUBLISHER_NAME = 'Publisher Name'
BULK_COL_ACCOUNT_PUBLISHER = 'Publisher'

#outputDf[BULK_COL_MARKET] = "<<YOUR VALUE>>"

TMP_MediaType = 'Unknown Media Type'
TMP_SubType = 'Unknown Media sub-type'
# blank out tmp field
inputDf[TMP_MediaType] = numpy.nan
inputDf[TMP_SubType] = numpy.nan

today = datetime.datetime.now(CLIENT_TIMEZONE).date()
print(tableize(inputDf))


inputDf.loc[ (inputDf[RPT_COL_ACCOUNT_PUBLISHER_NAME].str.contains('Taboola', na=False)) , TMP_MediaType ] = 'Native'
inputDf.loc[ (inputDf[RPT_COL_ACCOUNT_PUBLISHER_NAME].str.contains('Taboola', na=False)) , TMP_SubType ] = 'Native'
inputDf.loc[ (inputDf[RPT_COL_ACCOUNT_PUBLISHER_NAME].str.contains('TikTok', na=False)) , TMP_MediaType ] = 'Social'
inputDf.loc[ (inputDf[RPT_COL_ACCOUNT_PUBLISHER_NAME].str.contains('TikTok', na=False)) , TMP_SubType ] = 'TikTok'
inputDf.loc[ (inputDf[RPT_COL_ACCOUNT_PUBLISHER_NAME].str.contains('Facebook', na=False)) , TMP_MediaType ] = 'Social'
inputDf.loc[ (inputDf[RPT_COL_ACCOUNT_PUBLISHER_NAME].str.contains('Facebook', na=False)) , TMP_SubType ] = 'Meta'
inputDf.loc[ (inputDf[RPT_COL_ACCOUNT_PUBLISHER_NAME].str.contains('BING', na=False)) , TMP_MediaType ] = 'SEM'
inputDf.loc[ (inputDf[RPT_COL_ACCOUNT_PUBLISHER_NAME].str.contains('BING', na=False)) , TMP_SubType ] = 'SEM-Bing'
inputDf.loc[ (inputDf[RPT_COL_ACCOUNT_PUBLISHER_NAME].str.contains('Google', na=False)) , TMP_MediaType ] = 'SEM'
inputDf.loc[ (inputDf[RPT_COL_ACCOUNT_PUBLISHER_NAME].str.contains('Google', na=False)) , TMP_SubType ] = 'SEM-Google'
inputDf.loc[ (inputDf[RPT_COL_ACCOUNT_PUBLISHER_NAME].str.contains('Google', na=False)) & (inputDf[RPT_COL_CAMPAIGN].str.contains('\[GDN\]', na=False)), TMP_MediaType ] = 'Display'
inputDf.loc[ (inputDf[RPT_COL_ACCOUNT_PUBLISHER_NAME].str.contains('Google', na=False)) & (inputDf[RPT_COL_CAMPAIGN].str.contains('\[GDN\]', na=False)), TMP_SubType ] = 'Standard display'
inputDf.loc[ (inputDf[RPT_COL_ACCOUNT_PUBLISHER_NAME].str.contains('Google', na=False)) & (inputDf[RPT_COL_CAMPAIGN].str.contains('\[Demand Gen\]', na=False)), TMP_MediaType ] = 'Display & Online video'
inputDf.loc[ (inputDf[RPT_COL_ACCOUNT_PUBLISHER_NAME].str.contains('Google', na=False)) & (inputDf[RPT_COL_CAMPAIGN].str.contains('\[Demand Gen\]', na=False)), TMP_SubType ] = 'Demand Gen'
inputDf.loc[ (inputDf[RPT_COL_ACCOUNT_PUBLISHER_NAME].str.contains('Google', na=False)) & (inputDf[RPT_COL_CAMPAIGN].str.contains('\[YT\]', na=False)), TMP_MediaType ] = 'Online video'
inputDf.loc[ (inputDf[RPT_COL_ACCOUNT_PUBLISHER_NAME].str.contains('Google', na=False)) & (inputDf[RPT_COL_CAMPAIGN].str.contains('\[YT\]', na=False)), TMP_SubType ] = 'VIDEO-Instream'
inputDf.loc[ (inputDf[RPT_COL_ACCOUNT_PUBLISHER_NAME].str.contains('Google', na=False)) & (inputDf[RPT_COL_CAMPAIGN].str.contains('\[YT\]', na=False)) & (inputDf[RPT_COL_CAMPAIGN].str.contains('Bumper', na=False)), TMP_SubType ] = 'VIDEO-Bumper'
inputDf.loc[ (inputDf[RPT_COL_ACCOUNT].str.contains('Adform', na=False)) & (inputDf[RPT_COL_GROUP].str.contains('\[DISPLAY\]', na=False)), TMP_MediaType ] = 'Display'
inputDf.loc[ (inputDf[RPT_COL_ACCOUNT].str.contains('Adform', na=False)) & (inputDf[RPT_COL_GROUP].str.contains('\[DISPLAY\]', na=False)), TMP_SubType ] = 'Standard display'
inputDf.loc[ (inputDf[RPT_COL_ACCOUNT].str.contains('Adform', na=False)) & (inputDf[RPT_COL_GROUP].str.contains('\[DISPLAY-Premium\]', na=False)), TMP_MediaType ] = 'Display'
inputDf.loc[ (inputDf[RPT_COL_ACCOUNT].str.contains('Adform', na=False)) & (inputDf[RPT_COL_GROUP].str.contains('\[DISPLAY-Premium\]', na=False)), TMP_SubType ] = 'Premium display'
inputDf.loc[ (inputDf[RPT_COL_ACCOUNT].str.contains('Adform', na=False)) & (inputDf[RPT_COL_GROUP].str.contains('\[Email-Newsletter\]', na=False)), TMP_MediaType ] = 'Email'
inputDf.loc[ (inputDf[RPT_COL_ACCOUNT].str.contains('Adform', na=False)) & (inputDf[RPT_COL_GROUP].str.contains('\[Email-Newsletter\]', na=False)), TMP_SubType ] = 'Email-Newsletter'
inputDf.loc[ (inputDf[RPT_COL_ACCOUNT].str.contains('Adform', na=False)) & (inputDf[RPT_COL_GROUP].str.contains('\[Native\]', na=False)), TMP_MediaType ] = 'Native'
inputDf.loc[ (inputDf[RPT_COL_ACCOUNT].str.contains('Adform', na=False)) & (inputDf[RPT_COL_GROUP].str.contains('\[Native\]', na=False)), TMP_SubType ] = 'Native'
inputDf.loc[ (inputDf[RPT_COL_ACCOUNT].str.contains('Adform', na=False)) & (inputDf[RPT_COL_GROUP].str.contains('\[AUDIO-Podcast\]', na=False)), TMP_MediaType ] = 'Online Audio'
inputDf.loc[ (inputDf[RPT_COL_ACCOUNT].str.contains('Adform', na=False)) & (inputDf[RPT_COL_GROUP].str.contains('\[AUDIO-Podcast\]', na=False)), TMP_SubType ] = 'Audio-Podcast'
inputDf.loc[ (inputDf[RPT_COL_ACCOUNT].str.contains('Adform', na=False)) & (inputDf[RPT_COL_GROUP].str.contains('\[VIDEO-Instream\]', na=False)), TMP_MediaType ] = 'Online Video'
inputDf.loc[ (inputDf[RPT_COL_ACCOUNT].str.contains('Adform', na=False)) & (inputDf[RPT_COL_GROUP].str.contains('\[VIDEO-Instream\]', na=False)), TMP_SubType ] = 'VIDEO-Instream'
inputDf.loc[ (inputDf[RPT_COL_ACCOUNT].str.contains('Adform', na=False)) & (inputDf[RPT_COL_GROUP].str.contains('\[VIDEO-Bumper\]', na=False)), TMP_MediaType ] = 'Online Video'
inputDf.loc[ (inputDf[RPT_COL_ACCOUNT].str.contains('Adform', na=False)) & (inputDf[RPT_COL_GROUP].str.contains('\[VIDEO-Bumper\]', na=False)), TMP_SubType ] = 'VIDEO-Bumper'
inputDf.loc[ (inputDf[RPT_COL_ACCOUNT].str.contains('Offline', na=False)) & (inputDf[RPT_COL_CAMPAIGN].str.contains('OOH', na=False)), TMP_MediaType ] = 'OOH'
inputDf.loc[ (inputDf[RPT_COL_ACCOUNT].str.contains('Offline', na=False)) & (inputDf[RPT_COL_CAMPAIGN].str.contains('OOH', na=False)), TMP_SubType ] = 'OOH'
inputDf.loc[ (inputDf[RPT_COL_ACCOUNT].str.contains('Offline', na=False)) & (inputDf[RPT_COL_CAMPAIGN].str.contains('Radio', na=False)), TMP_MediaType ] = 'Radio'
inputDf.loc[ (inputDf[RPT_COL_ACCOUNT].str.contains('Offline', na=False)) & (inputDf[RPT_COL_CAMPAIGN].str.contains('Radio', na=False)), TMP_SubType ] = 'Radio'
inputDf.loc[ (inputDf[RPT_COL_ACCOUNT].str.contains('Offline', na=False)) & (inputDf[RPT_COL_CAMPAIGN].str.contains('TV', na=False)), TMP_MediaType ] = 'TV'
inputDf.loc[ (inputDf[RPT_COL_ACCOUNT].str.contains('Offline', na=False)) & (inputDf[RPT_COL_CAMPAIGN].str.contains('TV', na=False)), TMP_SubType ] = 'TV'
inputDf.loc[ (inputDf[RPT_COL_ACCOUNT].str.contains('Offline', na=False)) & (inputDf[RPT_COL_CAMPAIGN].str.contains('Print', na=False)), TMP_MediaType ] = 'Print'
inputDf.loc[ (inputDf[RPT_COL_ACCOUNT].str.contains('Offline', na=False)) & (inputDf[RPT_COL_CAMPAIGN].str.contains('Print', na=False)), TMP_SubType ] = 'Print'
inputDf.loc[ (inputDf[RPT_COL_ACCOUNT].str.contains('Offline', na=False)) & (inputDf[RPT_COL_GROUP].str.contains('\[Email-Newsletter\]', na=False)), TMP_MediaType ] = 'Email'
inputDf.loc[ (inputDf[RPT_COL_ACCOUNT].str.contains('Offline', na=False)) & (inputDf[RPT_COL_GROUP].str.contains('\[Email-Newsletter\]', na=False)), TMP_SubType ] = 'Email-Newsletter'
inputDf.loc[ (inputDf[RPT_COL_ACCOUNT].str.contains('Offline', na=False)) & (inputDf[RPT_COL_GROUP].str.contains('\[Email-Standalone\]', na=False)), TMP_MediaType ] = 'Email'
inputDf.loc[ (inputDf[RPT_COL_ACCOUNT].str.contains('Offline', na=False)) & (inputDf[RPT_COL_GROUP].str.contains('\[Email-Standalone\]', na=False)), TMP_SubType ] = 'Email-Standalone'
inputDf.loc[ (inputDf[RPT_COL_ACCOUNT].str.contains('Offline', na=False)) & (inputDf[RPT_COL_GROUP].str.contains('\[SMS\]', na=False)), TMP_MediaType ] = 'SMS'
inputDf.loc[ (inputDf[RPT_COL_ACCOUNT].str.contains('Offline', na=False)) & (inputDf[RPT_COL_GROUP].str.contains('\[SMS\]', na=False)), TMP_SubType ] = 'SMS'
inputDf.loc[ (inputDf[RPT_COL_ACCOUNT].str.contains('Offline', na=False)) & (inputDf[RPT_COL_GROUP].str.contains('\[Quiz\]', na=False)), TMP_MediaType ] = 'Quiz'
inputDf.loc[ (inputDf[RPT_COL_ACCOUNT].str.contains('Offline', na=False)) & (inputDf[RPT_COL_GROUP].str.contains('\[Quiz\]', na=False)), TMP_SubType ] = 'Quiz-Primetime'
inputDf.loc[ (inputDf[RPT_COL_ACCOUNT].str.contains('Email', na=False)) & (inputDf[RPT_COL_GROUP].str.contains('Affiliate', na=False)), TMP_MediaType ] = 'Affiliate'
inputDf.loc[ (inputDf[RPT_COL_ACCOUNT].str.contains('Email', na=False)) & (inputDf[RPT_COL_GROUP].str.contains('\[Affiliate]', na=False)), TMP_SubType ] = 'Affiliate'
inputDf.loc[ (inputDf[RPT_COL_ACCOUNT].str.contains('Email', na=False)) & (inputDf[RPT_COL_GROUP].str.contains('\[Affiliate-cashback]', na=False)), TMP_SubType ] = 'Affiliate-cashback'
inputDf.loc[ (inputDf[RPT_COL_ACCOUNT].str.contains('Email', na=False)) & (inputDf[RPT_COL_GROUP].str.contains('\[Affiliate-campaign]', na=False)), TMP_SubType ] = 'Affiliate-campaign'
inputDf.loc[ (inputDf[RPT_COL_ACCOUNT].str.contains('Email', na=False)) & (inputDf[RPT_COL_GROUP].str.contains('\[Affiliate-partner]', na=False)), TMP_SubType ] = 'Affiliate-partner'
print(tableize(inputDf))

print(inputDf.index.duplicated())

# copy new strategy to output
outputDf.loc[:,BULK_COL_MEDIA_TYPE] = inputDf.loc[:, TMP_MediaType]
outputDf.loc[:,BULK_COL_MEDIA_SUBTYPE] = inputDf.loc[:, TMP_SubType]

outputDf = outputDf[inputDf[TMP_MediaType].notnull() & inputDf[TMP_SubType].notnull() & ~inputDf[BULK_COL_CAMPAIGN].str.contains('"')]
outputDf = outputDf.drop_duplicates()
print(tableize(outputDf))

Post generated on 2024-05-15 07:44:05 GMT

comments powered by Disqus