Generating a supplemental description of an entity

1. A method comprising:

accessing multiple descriptions that describe multiple users associated with a target entity by a database of entity-user associations, the multiple descriptions each referencing the target entity, a portion of the multiple descriptions each including a token descriptive of the target entity and of a subset of the multiple users, the target entity being described by a primary description from which the token is absent;

determining that the token descriptive of the target entity and of the subset of the multiple users is overrepresented among tokens within the multiple descriptions by performing a statistical analysis of the tokens that includes:

generating a first probability model that the token occurs within the multiple descriptions of the multiple users associated with the target entity;

accessing a second probability model that the token occurs within all descriptions of all users represented in the database of entity-user associations; and

determining a statistical divergence between the first and second probability models of the token,

the determining that the token is overrepresented being performed by a processor of a machine based on the accessed multiple descriptions of the multiple users associated with the target entity;

generating a supplemental description of the target entity associated with the multiple users, the supplemental description including the overrepresented token determined based on the accessed multiple descriptions of the multiple users.

2. The method of claim 1 , wherein:

the primary description of the target entity describes the target entity as a business; and

the multiple descriptions describe the multiple users as employees of the business.

3. The method of claim 1 , wherein:

the primary description of the target entity describes the target entity as a first user within a social network; and

the multiple descriptions describe multiple users as further users of the social network that are connected to the first user by the social network.

4. The method of claim 1 , wherein:

the primary description of the target entity describes the target entity as a group of users within a social network; and

the multiple descriptions describe the multiple users as members of the group of users within the social network.

5. The method of claim 1 , wherein:

the primary description of the target entity describes the target entity as an event scheduled within a calendar database; and

the multiple descriptions describe the multiple users as attendees of the event scheduled within the calendar database.

6. The method of claim 1 , wherein:

the primary description of the target entity describes the target entity as a job that is available at a business; and

the multiple descriptions describe the multiple users as applicants to the job that is available at the business.

7. The method of claim 1 , wherein:

the primary description of the target entity describes the target entity as a news article; and

the multiple descriptions describe the multiple users as viewers of the news article.

8. The method of claim 1 , wherein:

the primary description of the target entity describes the target entity as a discussion thread in an online forum; and

the multiple descriptions describe the multiple users as participants in the discussion thread.

9. The method of claim 1 , further comprising:

receiving a plurality of indications of voluntary decisions by the multiple users to be voluntarily associated with the target entity; and

modifying the multiple descriptions of the multiple users based on the received plurality of indications of voluntary decisions,

the modified multiple descriptions describing the multiple users as voluntarily associated with the target entity by the voluntary decisions.

10. The method of claim 9 , wherein:

the modified multiple descriptions includes a profile of a first user among the multiple users, the profile describing the first user as being at least one of:

an employee of a business embodied by the target entity,

a connection of the target entity within a social network,

a member of a group embodied by the target entity,

an attendee of an event embodied by the target entity,

an applicant to a job embodied by the target entity,

a viewer of a news article embodied by the target entity, or

a participant in a discussion thread embodied by the target entity.

11. The method of claim 1 further comprising:

presenting a search result that references the target entity based on the generated supplemental description that describes the target entity and includes the overrepresented token descriptive of the target entity and of the subset of the multiple users.

12. The method of claim 1 further comprising:

ranking a search result that references the target entity based on the generated supplemental description that describes the target entity and includes the overrepresented token descriptive of the target entity and of the subset of the multiple users.

13. The method of claim 1 further comprising:

providing a recommendation of the target entity based on the generated supplemental description that describes the target entity and includes the overrepresented token descriptive of the target entity and of the subset of the multiple users.

14. The method of claim 1 further comprising:

providing a suggestion that an author of the primary description of the target entity modify the primary description based on the overrepresented token descriptive of the target entity and of the subset of the multiple users.

15. A non-transitory machine-readable storage medium comprising instructions that, when executed by one or more processors of a machine, cause the machine to perform operations comprising:

accessing multiple descriptions that describe multiple users associated with a target entity by a database of entity-user associations, the multiple descriptions each referencing the target entity, a portion of the multiple descriptions each including a token descriptive of the target entity and of a subset of the multiple users, the target entity being described by a primary description from which the token is absent;

determining that the token descriptive of the target entity and of the subset of the multiple users is overrepresented among tokens within the multiple descriptions by performing a statistical analysis of the tokens that includes:

generating a first probability model that the token occurs within the multiple descriptions of the multiple users associated with the target entity;

accessing a second probability model that the token occurs within all descriptions of all users represented in the database of entity-user associations; and

determining a statistical divergence between the first and second probability models of the token,

the determining that the token is overrepresented being performed by the one or more processors of the machine based on the accessed multiple descriptions of the multiple users associated with the target entity;

generating a supplemental description of the target entity associated with the multiple users, the supplemental description including the overrepresented token determined based on the accessed multiple descriptions of the multiple users.

16. The non-transitory machine-readable storage medium of claim 15 , wherein:

the overrepresented token is indicative of at least one of:

a skill possessed by each of the subset of the multiple users and utilized by the target entity,

a location of the subset of the multiple users and of the target entity,

a job title held by the subset of the multiple users and utilized by the target entity,

a school from which the subset of the multiple users graduated, or

an industry in which the subset of the multiple users work and to which the target entity pertains.

17. A system comprising:

an access module configured to access multiple descriptions that describe multiple users associated with a target entity by a database of entity-user associations, the multiple descriptions each referencing the target entity, a portion of the multiple descriptions each including a token descriptive of the target entity and of a subset of the multiple users, the target entity being described by a primary description from which the token is absent;

a processor configured by a determination module that configures the processor to determine that the token descriptive of the target entity and of the subset of the multiple users is overrepresented among a plurality of tokens within the multiple descriptions by performing a statistical analysis of the tokens that includes:

generating a first probability model that the token occurs within the multiple descriptions of the multiple users associated with the target entity;

accessing a second probability model that the token occurs within all descriptions of all users represented in the database of entity-user associations; and

determining a statistical divergence between the first and second probability models of the token,

the determining being based on the accessed multiple descriptions of the multiple users associated with the target entity;

a generation module configured to generate a supplemental description of the target entity associated with the multiple users, the supplemental description including the overrepresented token determined based on the accessed multiple descriptions of the multiple users.

18. The system of claim 17 , wherein:

the access module is configured to access the multiple descriptions of the multiple users in response to a search request for which the multiple descriptions are returned as search results,

the overrepresented token being absent from the search request to which the accessing of the multiple descriptions of the multiple users is responsive.