Automated Audience Selection Using Labeled Content Campaign Characteristics
An online system generates target audiences for a new content campaign based on target audiences for similar content campaigns. A content provider submits to the online system a request to create a new content campaign, and the online system compares keywords associated with the new campaign with keywords associated with existing campaigns to calculate and adjust similarity scores. The online system retrieves target audiences for campaigns that have a similarity score that equals or exceeds a threshold and uses the retrieved target audiences to create a target audience for the new campaign.
1 . A method comprising:
receiving, by an online system from a client device, a request to generate a new content campaign for a content provider on the online system;
determining, by the online system, keywords associated with the new content campaign by:
receiving at least one keyword from the content provider about the campaign, and
extracting at least one keyword from information provided to the online system by the content provider;
comparing the determined keywords for the new content campaign with keywords stored for other content campaigns that were previously determined by the online system for the other content campaigns;
identifying one or more of the other content campaigns as matching content campaigns based on a threshold level of match between stored keywords for the other campaigns and the determined keywords for the new content campaign;
identifying, for each of the matching content campaigns, a target audience of users of the online system to whom the matching content campaign targets content items;
generating a target audience for the new content campaign based at least in part on the target audiences of the identified matching content campaigns; and
presenting the target audience for the new content campaign to the client device as a recommendation by the online system for the new content campaign.
2 . The method of claim 1 , wherein extracting at least one keyword comprises extracting the keyword by the online system from sponsored content associated with the new content campaign.
3 . The method of claim 1 , wherein extracting at least one keyword comprises extracting the keyword by the online system from an online system profile about the content provider.
4 . The method of claim 1 , wherein extracting at least one keyword comprises extracting the keyword by the online system from a brand page within the online system that is associated with the content provider's profile.
5 . The method of claim 1 , wherein extracting at least one keyword comprises extracting the keyword by the online system from a third-party website that is separate from the online system and that is affiliated with the content provider.
6 . The method of claim 1 , further comprising ranking the keywords based on the number of times that each of the keywords appears in the information.
7 . The method of claim 1 , further comprising ranking the keywords based on the placement of each of the keywords in the information.
8 . The method of claim 1 , further comprising ranking the keywords based on how conspicuous the keywords are relative to surrounding text.
9 . The method of claim 1 , wherein identifying one or more of the content campaigns as matching content campaigns comprises calculating a similarity score above a threshold level of match.
10 . The method of claim 9 , further comprising adjusting the similarity score based on the frequency of occurrence of the keywords.
11 . The method of claim 9 , further comprising adjusting the similarity score based on the popularity of the keywords.
12 . The method of claim 9 , further comprising adjusting the similarity score based on a ranking of keywords provided by the content provider or performed by the online system.
13 . The method of claim 1 , wherein generating a target audience for the new content campaign further comprises extracting an overlapping subset of users targeted by two or more content campaigns.
14 . The method of claim 13 , further comprising including the overlapping subset of users at a higher proportion than non-overlapping users.
15 . The method of claim 1 , further comprising calculating a bid amount or adjusting a bid amount associated with the content campaign in response to the online system labeling another content provider as similar to the content provider associated with the content campaign.
16 . A non-transitory computer readable storage medium storing instructions that, when executed by a computing system, cause the computing system to perform operations comprising:
receiving, by an online system from a client device, a request to generate a new content campaign for a content provider on the online system;
determining, by the online system, keywords associated with the new content campaign by:
receiving at least one keyword from the content provider about the campaign, and
extracting at least one keyword from information provided to the online system by the content provider;
comparing the determined keywords for the new content campaign with keywords stored for other content campaigns that were previously determined by the online system for the other content campaigns;
identifying one or more of the other content campaigns as matching content campaigns based on a threshold level of match between stored keywords for the other campaigns and the determined keywords for the new content campaign;
identifying, for each of the matching content campaigns, a target audience of users of the online system to whom the matching content campaign targets content items;
generating a target audience for the new content campaign based at least in part on the target audiences of the identified matching content campaigns; and
presenting the target audience for the new content campaign to the client device as a recommendation by the online system for the new content campaign.
17 . The non-transitory computer readable storage medium of claim 16 , wherein extracting at least one keyword comprises extracting the keyword by the online system from sponsored content associated with the new content campaign.
18 . The non-transitory computer readable storage medium of claim 16 , wherein extracting at least one keyword comprises extracting the keyword by the online system from an online system profile about the content provider.
19 . The non-transitory computer readable storage medium of claim 16 , wherein identifying one or more of the content campaigns as matching content campaigns comprises calculating a similarity score above a threshold of match.
20 . The non-transitory computer readable storage medium of claim 16 , wherein generating a target audience for the new content campaign further comprises extracting an overlapping subset of users targeted by two or more content campaigns.