Content item selection criteria generation

Abstract

Selection of content selection criteria based on entities related by relationship dimensions. In one aspect, a method receives a selection of a seed entity described in entity relation data, the entity relation data defining instances of entities, and for each entity one or more relationship dimensions; generating a set of selected entities; iteratively updating the set of selected entities, each iteration comprising: determining a set of relationship dimensions from the entities in the set of selected entities, each relationship dimension in the set being selected from the one or more relationship dimensions of the entities in the set of selected entities, receiving a selection of one of the relationship dimensions and in response: determining a set of candidate entities from the relationship dimensions and in response to receiving a selection of one or more candidate entities, updating the set of selected entities to include the one or more candidate entities.

What is claimed is: 
     
         1 . A computer implemented method, comprising:
 receiving a selection of a seed entity described in entity relation data, wherein the entity relation data defines instances of entities, and for each entity one or more relationship dimensions, each relationship dimension defining a relationship between the entity and one or more other entities;   generating a set of selected entities, the set of selected entities being the seed entity;   iteratively updating the set of selected entities, each iteration comprising:
 determining a set of relationship dimensions from the entities in the set of selected entities, each relationship dimension in the set being selected from the one or more relationship dimensions of the entities in the set of selected entities; 
 receiving a selection of one of the relationship dimensions and in response: 
 determining a set of candidate entities, each candidate entity in the set being an entity related to one of the entities in the set of selected entities by selected relationship dimension; and 
 in response to receiving a selection of one or more candidate entities, updating the set of selected entities to include the one or more candidate entities. 
   
     
     
         2 . The computer implemented method of  claim 1 , further comprising:
 receiving a request to generate content selection criteria based on the set of selected entities, the content selection criteria being criteria for selecting content to be provided to a user device, and in response:
 providing the set of selected entities to a content selection criteria generator; 
 receiving, from the content selection criteria generators, a set of candidate content selection criterion based on the set of selected entities; 
 receiving selections of a subset of the candidate content selection criterion; and 
 storing the selected candidate content selection criterion as content selection criteria for use by a content management system to select and provide content to user devices in accordance with the content selection criteria. 
   
     
     
         3 . The computer implemented method of  claim 2 , wherein:
 receiving the set of candidate content selection criterion comprises receiving a set of keywords based on the set of selected entities; and   storing the selected candidate content selection criterion as content selection criteria for use by a content management system comprises storing the selected keywords in advertising campaign data for use in selecting and providing advertisements to user devices.   
     
     
         4 . The computer implemented method of  claim 1 , wherein the entity relation data comprises data defining a knowledge graph having a plurality of nodes and edges, wherein each node in the knowledge graph represents a different entity and pairs of nodes in the knowledge graph are connected by one or more edges, each edge representing a relationship dimension that defines a relationship between the two entities represented by the pair of nodes. 
     
     
         5 . The computer implemented method of  claim 4 , wherein determining the set of candidate entities comprise determining entities that are within N nodes of a node representing an entity connected by an edge representing the selected relationship dimension to another node representing one of the entities in the set of candidate entities. 
     
     
         6 . The computer implemented method of  claim 5 , where N is greater than 0. 
     
     
         7 . The computer implemented method of  claim 2 , wherein:
 the entity relation data comprises data defining instances of entities, and for each entity a plurality of attributes of the entity;   determining a set of relationship dimensions comprises determining a set of attributes from the entities in the set of selected entities; and   determining a set of candidate entities comprising determining a set of entities not included in the set of selected entities and that each have at least one attribute in the determined set of attributes.   
     
     
         8 . The computer implemented method of  claim 2 , wherein:
 receiving a selection of one of the relationship dimensions comprises receiving a selection of a first relationship dimension indicating a positive relationship dimension; and   further comprising receiving a selection of a second relationship dimension indicating a negative relationship dimension;   wherein determining a set of candidate entities comprises determine candidate entities that are entity related to one of the entities in the set of selected entities by the positive relationship dimension and not related to any entities in the set of selected entities by the negative relationship dimension.   
     
     
         9 . A system, comprising:
 a data processing apparatus;   a data store storing instructions executable by the data processing and that upon execution cause the data processing apparatus to perform operations comprising:
 receiving a selection of a seed entity described in entity relation data, wherein the entity relation data defines instances of entities, and for each entity one or more relationship dimensions, each relationship dimension defining a relationship between the entity and one or more other entities; 
 generating a set of selected entities, the set of selected entities being the seed entity; 
 iteratively updating the set of selected entities, each iteration comprising:
 determining a set of relationship dimensions from the entities in the set of selected entities, each relationship dimension in the set being selected from the one or more relationship dimensions of the entities in the set of selected entities; 
 receiving a selection of one of the relationship dimensions and in response: 
 determining a set of candidate entities, each candidate entity in the set being an entity related to one of the entities in the set of selected entities by selected relationship dimension; and 
 in response to receiving a selection of one or more candidate entities, updating the set of selected entities to include the one or more candidate entities. 
 
   
     
     
         10 . The system of  claim 9 , wherein the operations performed by the data processing apparatus further comprise:
 receiving a request to generate content selection criteria based on the set of selected entities, the content selection criteria being criteria for selecting content to be provided to a user device, and in response:
 providing the set of selected entities to a content selection criteria generator; 
 receiving, from the content selection criteria generators, a set of candidate content selection criterion based on the set of selected entities; 
 receiving selections of a subset of the candidate content selection criterion; and 
 storing the selected candidate content selection criterion as content selection criteria for use by a content management system to select and provide content to user devices in accordance with the content selection criteria. 
   
     
     
         11 . The system of  claim 10 , wherein:
 the operation of receiving the set of candidate content selection criterion comprises receiving a set of keywords based on the set of selected entities; and   the operation of storing the selected candidate content selection criterion as content selection criteria for use by a content management system comprises storing the selected keywords in advertising campaign data for use in selecting and providing advertisements to user devices.   
     
     
         12 . The system of  claim 9 , wherein the entity relation data comprises data defining a knowledge graph having a plurality of nodes and edges, wherein each node in the knowledge graph represents a different entity and pairs of nodes in the knowledge graph are connected by one or more edges, each edge representing a relationship dimension that defines a relationship between the two entities represented by the pair of nodes. 
     
     
         13 . The system of  claim 12 , wherein the operation of determining the set of candidate entities comprise determining entities that are within N nodes of a node representing an entity connected by an edge representing the selected relationship dimension to another node representing one of the entities in the set of candidate entities. 
     
     
         14 . The system of  claim 13 , where N is greater than 0. 
     
     
         15 . The system of  claim 10 , wherein:
 the entity relation data comprises data defining instances of entities, and for each entity a plurality of attributes of the entity;   the operation of determining a set of relationship dimensions comprises determining a set of attributes from the entities in the set of selected entities; and   the operation of determining a set of candidate entities comprising determining a set of entities not included in the set of selected entities and that each have at least one attribute in the determined set of attributes.   
     
     
         16 . The system of  claim 10 , wherein:
 the operation of receiving a selection of one of the relationship dimensions comprises receiving a selection of a first relationship dimension indicating a positive relationship dimension; and   the operations performed by the data processing apparatus further comprise receiving a selection of a second relationship dimension indicating a negative relationship dimension;   wherein the operation of determining a set of candidate entities comprises determine candidate entities that are entity related to one of the entities in the set of selected entities by the positive relationship dimension and not related to any entities in the set of selected entities by the negative relationship dimension.   
     
     
         17 . A computer storage medium encoded with instructions that when executed by a data processing apparatus cause the one or more data processing apparatus to perform operations comprising:
 receiving a selection of a seed entity described in entity relation data, wherein the entity relation data defines instances of entities, and for each entity one or more relationship dimensions, each relationship dimension defining a relationship between the entity and one or more other entities;   generating a set of selected entities, the set of selected entities being the seed entity;   iteratively updating the set of selected entities, each iteration comprising:
 determining a set of relationship dimensions from the entities in the set of selected entities, each relationship dimension in the set being selected from the one or more relationship dimensions of the entities in the set of selected entities; 
 receiving a selection of one of the relationship dimensions and in response: 
 determining a set of candidate entities, each candidate entity in the set being an entity related to one of the entities in the set of selected entities by selected relationship dimension; and 
 in response to receiving a selection of one or more candidate entities, updating the set of selected entities to include the one or more candidate entities. 
   
     
     
         18 . The computer storage medium of  claim 17 , wherein the instructions cause the data processing apparatus to perform further operations comprising:
 receiving a request to generate content selection criteria based on the set of selected entities, the content selection criteria being criteria for selecting content to be provided to a user device, and in response:
 providing the set of selected entities to a content selection criteria generator; 
 receiving, from the content selection criteria generators, a set of candidate content selection criterion based on the set of selected entities; 
 receiving selections of a subset of the candidate content selection criterion; and 
 storing the selected candidate content selection criterion as content selection criteria for use by a content management system to select and provide content to user devices in accordance with the content selection criteria.