Back to results
US 9,805,126 B2Granted
Context-based ranking of search results
AssigneeFacebook, Inc.
InventorsRyan Patterson, Michael Dudley Johnson, Erick Tseng
Life of this patent
Estimating the term…
Abstract
In one embodiment, a computing device may access a search query provided by a user; identify a set of search results in response to the search query, wherein one or more search results in the set are associated with a feature of a social-networking system; rank the set of search results based on one or more factors; boost one or more ranks of the one or more search results associated with the feature to bring the feature to the user's attention; and present the set of search results to the user in order of its ranking.
Claims18
A method comprising, by one or more computer systems of a social-networking system:
- receiving, from a client device of a first user of the social-networking system, a search query comprising one or more characters inputted by the first user, the one or more characters being received as the first user inputs the one or more characters at the client device into a user interface of the social-networking system;
- identifying one or more entities associated with the social-networking system matching the one or more characters of the search query;
- ranking each of the identified entities matching the search query based on a calculated likelihood that the user will interact with a search result corresponding to the identified entity, wherein the calculated likelihood that the user will interact with the search result corresponding to the identified entity is based at least in part on a historical behavior of the user in selecting one or more prior search results related to the identified entity;
- boosting one or more ranks of one or more of the identified entities matching the search query based on:
- one or more business objectives of the social-networking system, wherein at least one of the business objectives comprises encouraging the first user to interact with the identified entity based on a level of social relevance of the identified entity with respect to the first user;
- and
- a level of the user's interaction with one or more features of the social-networking system associated with the identified entity, wherein the boosting is designed to increase the level of the user's interaction with the one or more features of the social-networking system associated with the identified entity;
- and
- sending, to the client device of the first user and in response to the received search query, one or more search results for presentation to the user as the user inputs the one or more characters at the client device into the user interface, each search result corresponding to one of the identified entities, the search results being presented according to the ranking and boosting of the corresponding identified entities, the presentation of the search results to the user enabling the user to select one or more of the search results to interact with.
6 more elements
A system comprising:
- a memory;
- and
- one or more processors coupled to the memory and configured to perform operations comprising:
- receive, from a client device of a first user of a social-networking system, a search query comprising one or more characters inputted by the first user, the one or more characters being received as the first user inputs the one or more characters at the client device into a user interface of the social-networking system;
- identify one or more entities associated with the social-networking system matching the one or more characters of the search query;
- rank each of the identified entities matching the search query based on a calculated likelihood that the user will interact with a search result corresponding to the identified entity, wherein the calculated likelihood that the user will interact with the search result corresponding to the identified entity is based at least in part on a historical behavior of the user in selecting one or more prior search results related to the identified entity;
- boost one or more ranks of one or more of the identified entities matching the search query based on:
- one or more business objectives of the social-networking system, wherein at least one of the business objectives comprises encouraging the first user to interact with the identified entity based on a level of social relevance of the identified entity with respect to the first user;
- and
- a level of the user's interaction with one or more features of the social-networking system associated with the identified entity, wherein the boosting is designed to increase the level of the user's interaction with the one or more features of the social-networking system associated with the identified entity;
- and
- send, to the client device of the first user and in response to the received search query, one or more search results for presentation to the user as the user inputs the one or more characters at the client device into the user interface, each search result corresponding to one of the identified entities, the search results being presented according to the ranking and boosting of the corresponding identified entities, the presentation of the search results to the user enabling the user to select one or more of the search results to interact with.
9 more elements
One or more computer-readable non-transitory storage media embodying software operable when executed by one or more computer systems to:
- receive, from a client device of a first user of a social-networking system, a search query comprising one or more characters inputted by the first user, the one or more characters being received as the first user inputs the one or more characters at the client device into a user interface of the social-networking system;
- identify one or more entities associated with the social-networking system matching the one or more characters of the search query;
- rank each of the identified entities matching the search query based on a calculated likelihood that the user will interact with a search result corresponding to the identified entity, wherein the calculated likelihood that the user will interact with the search result corresponding to the identified entity is based at least in part on a historical behavior of the user in selecting one or more prior search results related to the identified entity;
- boost one or more ranks of one or more of the identified entities matching the search query based on:
- one or more business objectives of the social-networking system, wherein at least one of the business objectives comprises encouraging the first user to interact with the identified entity based on a level of social relevance of the identified entity with respect to the first user;
- and
- a level of the user's interaction with one or more features of the social-networking system associated with the identified entity, wherein the boosting is designed to increase the level of the user's interaction with the one or more features of the social-networking system associated with the identified entity;
- and
- send, to the client device of the first user and in response to the received search query, one or more search results for presentation to the user as the user inputs the one or more characters at the client device into the user interface, each search result corresponding to one of the identified entities, the search results being presented according to the ranking and boosting of the corresponding identified entities, the presentation of the search results to the user enabling the user to select one or more of the search results to interact with.
6 more elements
Dependencies are inferred from the claim text.