Related Entity Search

• US 20160063106A1
• Filed: 08/08/2012
• Published: 03/03/2016
• Est. Priority Date: 08/08/2012
• Status: Abandoned Application

• Current Assignee: Google LLC (Alphabet Inc.)
• Original Assignee: Google Inc. (Alphabet Inc.)
• Inventors: Huynh, David Francois, Lou, JinYu, Ding, Chen, Chai, Ying
• Application Number:
  US13/569,698
• Publication Number:
  US 20160063106A1
• Time in Patent Office: Days
• US Class Current: 1/1
• CPC Class Codes: G06F 16/9024 Graphs; Linked lists G06F16/9027 takes precedence
• CPC Class Codes: G06F 16/951 Indexing; Web crawling techniques

Abstract

Methods, systems, and computer-readable media are provided for related entity search. In some implementations, a search system identifies a collection of entities associated with a search query. The search system identifies entities related to those entities, and determines the relationships between them. The search system provides access to one or more collections of related results identifying entities based on the determined relationships.

Claims

1. A method comprising:

obtaining, using one or more computers, a plurality of search results identifying entities from a knowledge graph, wherein at least one search result of the plurality of search results is associated with an entity in the knowledge graph;

identifying, using one or more computers, one or more related entities from the knowledge graph, wherein each related entity is related to at least one of the entities associated with the plurality of search results;

determining, using one or more computers, one or more graph edges of the knowledge graph that link nodes corresponding to the one or more entities with nodes corresponding to the one or more related entities;

generating a triple for each of the one or more determined graph edges, the triple comprising a grouping of one of the one or more graph edges, a first node corresponding to one of the one or more entities, and a second node corresponding to one of the one or more related entities;

analyzing each generated triple to identify additional entities in the knowledge graph and a related plurality of results; and

providing, using one or more computers, access to the related plurality of results.

2. The method of claim 1, wherein providing access to the related plurality of results comprises providing access to a collection of results identifying entities that are all of the same type.

3. The method of claim 1, wherein providing access to the related plurality of results comprises providing links based on relevance.

5. The method of claim 1, wherein providing access to the related plurality of results comprises providing access to a collection of results that are substantially contained in the plurality of search results.

6. The method of claim 1, wherein determining the one or more graph edges comprises identifying one or more entity types associated with the entities and the related entities.

7. The method of claim 1, wherein determining the one or more graph edges comprises identifying values associated with the entities and the related entities.

8. The method of claim 1, wherein providing access to the related plurality of results comprises displaying links to collections in a related collection area of a display screen.

9. The method of claim 1, wherein providing access to the related plurality of results comprises displaying links in a horizontally scrollable display.

10. The method of claim 1, wherein providing access to the related plurality of results comprises displaying links in a pull-down menu.

31. The method of claim 1, further comprising:

identifying, based on at least on one generated triple, one or more alternative text queries;

determining additional results based on the one or more alternative text queries; and

providing access to the additional results.

4. The method of claim herein providing access to the related plurality of results comprises identifying results identifying entities having nodes that form a triple corresponding to a relationship matching at least one of a plurality of relationships.

11. A system comprising:

a database comprising a knowledge graph; and

one or more computers configured to perform operations including;

obtaining a plurality of search results identifying entities from the knowledge graph, wherein at least one search result of the plurality of search results identifying entities is associated with an entity in the knowledge graph,

identifying one or more related entities from the knowledge graph, wherein each related entity is related to at least one of the entities associated with the plurality of search results identifying entities,

determining one or more graph edges of the knowledge graph that link nodes corresponding to the one or more entities with nodes corresponding to the one or more related entities,

generating a triple for each of the one or more determined graph edges, the triple comprising a grouping of one of the one or more graph edges, a first node corresponding to one of the one or more entities, and a second node corresponding to one of the one or more related entities,

analyzing each generated triple to identify additional entities in the knowledge graph and a related plurality of results, and

providing access to the related plurality of results.

12. The system of claim 11, wherein the one or more computers are further configured to provide access to a collection of results identifying entities that are all of the same type.

13. The system of claim 11, wherein the one or more computers are further configured to provide links based on relevance.

14. The system of claim 11, wherein the one or more computers are further configured to identify results identifying entities having nodes that form a triple corresponding to a relationship matching at least one of a plurality of relationships.

15. The system of claim 11, wherein the one or more computers are further configured to provide access to a collection of results that are substantially contained in the plurality of search results.

16. The system of claim 11, wherein the one or more computers are further configured to identify one or more entity types associated with the entities and the related entities.

17. The system of claim 11, wherein the one or more computers are further configured to identify values associated with the entities and the related entities.

18. The system of claim 11, wherein the one or more computers are further configured to display links to collections in a related collection area of a display screen.

19. The system of claim 11, wherein the one or more computers are further configured to display links in a horizontally scrollable display.

20. The system of claim 11, wherein the one or more computers are further configured to display links in a pull-down menu.

32. The system of claim 11, the one or more computers are further configured to perform operations including:

identifying, based on at least in part on one generated triple, one or ore alternative text queries;

determining additional results based on the one or more alternative text queries; and

providing access to the additional results.

21. A non-transitory computer-readable medium having instructions recorded thereon that, when executed by at least one processor, cause the processor to perform a method comprising:

obtaining, using one or more computers, a plurality of search results identifying entities from a knowledge graph, wherein at least one search result of the plurality of search results identifying entities is associated with an entity in the knowledge graph;

identifying, using one or more computers, one or more related entities from the knowledge graph, wherein each related entity is related to at least one of the entities associated with the plurality of search results identifying entities;

determining, using one or more computers, one or more graph edges of the knowledge graph linking nodes corresponding to the one or more entities with nodes corresponding to the one or more related entities;

generating a triple for each of the one or more determined graph edges, the triple comprising a grouping of one of the one or more graph edges, a first node corresponding to one of the one or more entities, and a second node corresponding to one of the one or more related entities;

analyzing each generated triple to identify additional entities in the knowledge graph and a related plurality of results; and

providing, using one or more computers, access to the related plurality of results identifying entities.

22. The non-transitory computer readable medium of claim 21, wherein providing access to the related plurality of results comprises providing access to a collection of results identifying entities that are all of the same type.

23. The non-transitory computer readable medium of claim 21, wherein providing access to the related plurality of results comprises providing links based on relevance.

24. The non-transitory computer readable medium of claim 21, wherein providing access to the related plurality of results comprises identifying results identifying entities having nodes that form a triple corresponding to a relationship matching at least one of a plurality of relationships.

25. The non-transitory computer readable medium of claim 21, wherein providing access to the related plurality of results comprises providing access to a collection of results that are substantially contained in the plurality of search results.

26. The non-transitory computer readable medium of claim 21, wherein determining the one or more graph edges comprises identifying one or more entity types associated with the entities and the related entities.

27. The non-transitory computer readable medium of claim 21, wherein determining the one or more graph edges comprises identifying values associated with the entities and the related entities.

28. The non-transitory computer readable medium of claim 21, wherein providing access to the related plurality of results comprises displaying links to collections in a related collection area of a display screen.

29. The non-transitory computer readable medium of claim 21, wherein providing access to the related plurality of results comprises displaying links in a horizontally scrollable display.

30. The non-transitory computer readable medium of claim 21, wherein providing access to the related plurality of results comprises displaying links in a pull-down menu.

33. The non-transitory computer readable medium of claim 21, wherein the method further comprises:

identifying, based on at least in part on one generated triple, one or more alternative text queries;

determining additional results based on the one or more alternative text queries; and

providing access to the additional results.