Content item audience selection
Methods, systems, and apparatus, including computer programs encoded on a computer storage medium, for selecting an audience for content are disclosed. In one aspect, a method includes receiving, from a content item provider, a request to distribute content items to users that have been deemed interested in a particular entity. First users that have expressed an interest in the particular entity are identified. Expansion entities for the particular entity are identified in a knowledge graph. At least one of the expansion entities can be connected to the particular entity by a relationship path. Second users are identified. The second users are deemed interested in the particular entity based on the second users having expressed an interest in an expansion entity. The content items are provided to at least a portion of the first users and at least a portion of the second users.
1. A method performed by one or more data processing apparatus comprising one or more processors, the method comprising:
receiving a request to distribute content to a specified number of users of a particular social network that have expressed an interest in a first entity;
determining that the one or more data processing apparatus are incapable of distributing the content to the specified number of users because fewer than the specified number of users of the particular social network have expressed an interest in the first entity;
in response to determining that the one or more data processing apparatus are incapable of distributing the content to the specified number of users, creating an expanded audience for the content that includes first users that have expressed an interest in the first entity within the particular social network and additional users that have not expressed an interest in the first entity within the particular social network, including:
identifying a particular user that has expressed an interest in the first entity within the particular social network;
identifying a set of users that are socially connected to the particular user in the particular social network;
identifying, as candidate entities, entities in which at least one of the users in the set of users has shown an interest;
identifying, for each candidate entity, a relationship between the candidate entity and the first entity;
determining, for each candidate entity, a relationship score based on a number of users in the set of users that have shown an interest in other entities that have the same relationship with the first entity as the candidate entity;
selecting, from the candidate entities, one or more expansion entities based at least on each identified relationship and the relationship score for each candidate entity; and
selecting, as the additional users, users in the particular social network that have expressed an interest in at least one of the one or more expansion entities, but have not expressed an interest in the first entity; and
enabling distribution of the content to the specified number of users of the particular social network by distributing the content to the expanded audience that was expanded to include first users and the additional users in the expanded audience.
2. The method of claim 1 , further comprising determining that fewer than the specified number of users of the particular social network have expressed an express interest in the first entity by determining that fewer of than the specified number of users of the particular social network have interacted with a positive feedback element that references the first entity.
3. The method of claim 1 , wherein identifying, for each candidate entity, a relationship between the candidate entity and the first entity comprises identifying each relationship using a knowledge graph that includes nodes that represent entities and edges that indicate relationships between entities connected by the edges.
4. The method of claim 1 , wherein selecting, from the candidate entities, one or more expansion entities based at least on each identified relationship comprises selecting, as an expansion entity, each candidate entity having a same relationship with the first entity as each expansion entity selected based on each relationship score.
5. The method of claim 1 , further comprising:
obtaining a performance measure for each relationship based on a number of user interactions with the content that was provided to users that have expressed an interest in entities having the relationship with the first entity; and
updating a relationship score for the candidate entities based on the obtained performance measures.
6. A non-transitory computer storage medium encoded with a computer program, the program comprising instructions that when executed by one or more data processing apparatus cause the one or more data processing apparatus to perform operations comprising:
receiving a request to distribute content to a specified number of users of a particular social network that have expressed an interest in a first entity;
determining that the one or more data processing apparatus are incapable of distributing the content to the specified number of users because fewer than the specified number of users of the particular social network have expressed an interest in the first entity;
in response to determining that the one or more data processing apparatus are incapable of distributing the content to the specified number of users, creating an expanded audience for the content that includes first users that have expressed an interest in the first entity within the particular social network and additional users that have not expressed an interest in the first entity within the particular social network, including:
identifying a particular user that has expressed an interest in the first entity within the particular social network;
identifying a set of users that are socially connected to the particular user in the particular social network;
identifying, as candidate entities, entities in which at least one of the users in the set of users has shown an interest;
identifying, for each candidate entity, a relationship between the candidate entity and the first entity;
determining, for each candidate entity, a relationship score based on a number of users in the set of users that have shown an interest in other entities that have the same relationship with the first entity as the candidate entity;
selecting, from the candidate entities, one or more expansion entities based at least on each identified relationship and the relationship score for each candidate entity; and
selecting, as the additional users, users in the particular social network that have expressed an interest in at least one of the one or more expansion entities, but have not expressed an interest in the first entity; and
enabling distribution of the content to the specified number of users of the particular social network by distributing the content to the expanded audience that was expanded to include first users and the additional users.
7. The non-transitory computer storage medium of claim 6 , wherein the operations comprise determining that fewer than the specified number of users of the particular social network have expressed an express interest in the first entity by determining that fewer of than the specified number of users of the particular social network have interacted with a positive feedback element that references the first entity.
8. The non-transitory computer storage medium of claim 6 , wherein identifying, for each candidate entity, a relationship between the candidate entity and the first entity comprises identifying each relationship using a knowledge graph that includes nodes that represent entities and edges that indicate relationships between entities connected by the edges.
9. The non-transitory computer storage medium of claim 6 , wherein selecting, from the candidate entities, one or more expansion entities based at least on each identified relationship comprises selecting, as an expansion entity, each candidate entity having a same relationship with the first entity as each expansion entity selected based on each relationship score.
10. A system, comprising:
one or more processors; and
a computer-readable storage device coupled to the one or more processors and having instructions stored thereon which, when executed by the one or more processors, cause the one or more processors to perform operations, the operations comprising:
receiving a request to distribute content to a specified number of users of a particular social network that have expressed an interest in a first entity;
determining that the one or more data processing apparatus are incapable of distributing the content to the specified number of users because fewer than the specified number of users of the particular social network have expressed an interest in the first entity;
in response to determining that the one or more data processing apparatus are incapable of distributing the content to the specified number of users, creating an expanded audience for the content that includes first users that have expressed an interest in the first entity within the particular social network and additional users that have not expressed an interest in the first entity within the particular social network, including:
identifying a particular user that has expressed an interest in the first entity within the particular social network;
identifying a set of users that are socially connected to the particular user in the particular social network;
identifying, as candidate entities, entities in which at least one of the users in the set of users has shown an interest;
identifying, for each candidate entity, a relationship between the candidate entity and the first entity;
determining, for each candidate entity, a relationship score based on a number of users in the set of users that have shown an interest in other entities that have the same relationship with the first entity as the candidate entity;
selecting, from the candidate entities, one or more expansion entities based at least on each identified relationship and the relationship score for each candidate entity; and
selecting, as the additional users, users in the particular social network that have expressed an interest in at least one of the one or more expansion entities, but have not expressed an interest in the first entity; and
enabling distribution of the content to the specified number of users of the particular social network by distributing the content to the expanded audience that was expanded to include first users and the additional users.
11. The system of claim 10 , wherein the operations comprise determining that fewer than the specified number of users of the social network have expressed an express interest in the first entity by determining that fewer of than the specified number of users of the particular social network have interacted with a positive feedback element that references the first entity.
12. The system of claim 10 , wherein identifying, for each candidate entity, a relationship between the candidate entity and the first entity comprises identifying each relationship using a knowledge graph that includes nodes that represent entities and edges that indicate relationships between entities connected by the edges.
13. The system of claim 10 , wherein selecting, from the candidate entities, one or more expansion entities based at least on each identified relationship comprises selecting, as an expansion entity, each candidate entity having a same relationship with the first entity as each expansion entity selected based on each relationship score.
14. The system of claim 10 , wherein the operations comprise:
obtaining a performance measure for each relationship based on a number of user interactions with the content that was provided to users that have expressed an interest in entities having the relationship with the first entity; and
updating a relationship score for the candidate entities based on the obtained performance measures.