Search personalization and an enterprise knowledge graph

• US 9,998,472 B2
• Filed: 05/28/2015
• Issued: 06/12/2018
• Est. Priority Date: 05/28/2015
• Status: Active Grant

• Current Assignee: Google LLC (Alphabet Inc.)
• Original Assignee: Google LLC (Alphabet Inc.)
• Inventors: VerWeyst, Brent, Cochran, Martin James, Li, Jimin, Puig, Albert Orriols, Xiao, John
• Primary Examiner(s): Casanova, Jorge A
• Application Number:
  US14/723,827
• Publication Number:
  US 20160350307A1
• Time in Patent Office: 1,111 Days
• CPC Class Codes: G06F 21/6227 where protection concerns the structure of data, e.g. records, types, queries
• CPC Class Codes: H04L 63/101 Access control lists [ACL]

Abstract

Methods, systems, and apparatus, including computer programs encoded on a computer storage medium for searches associated with the enterprise information. In one aspect, a method includes receiving a query from a member of an enterprise; searching an index that includes member information and resources: the resources include entities and entity facts related to the enterprise; the entity facts are identified from the entities; the index includes data defining access privileges to the entities and the entity facts according entity fact access control lists; determining the entity facts that are accessible to the member according to the entity fact access control lists; determining, based on member information and entity facts, search result data including data describing entities and entity facts relevant to the query; and providing search results to the member of the enterprise, the search results including the entities and entity facts determined to be relevant to the query.

Claims

1. A method implemented in a data processing apparatus, comprising:

receiving a query from a member of an enterprise;

searching an index that includes member information of members of the enterprise and documents of the enterprise, wherein;

the documents include data describing entities and entity facts related to the enterprise and relationships among the entities, and each document has a respective access control list specifying access privileges to the document for members of the enterprise;

the entity facts are identified from the entities of the documents of the enterprise;

each entity fact describes at least one feature of the entity, wherein the feature of the entity is a relationship between the entity and another entity and wherein each entity fact is derived from one or more corresponding documents in which the entity fact is described; and

the index includes data defining access privileges to the data describing the entities and the entity facts according to respective entity fact access control lists, wherein each entity fact access control list is different from the access control lists provided for the documents of the enterprise, and each entity fact inherits an access control list of a document from which the entity fact is derived, wherein deriving the entity facts comprises selecting each document from the documents, and for the selected document;

determining a first entity identified within the document;

determining a second entity identified within the document;

determining a relationship between the first entity and the second entity that is described within the document; and

generating, as the entity fact, data describing the first entity, the second entity, and the relationship between the first entity and the second entity as described in the document;

wherein multiple entity facts are derived from a selected document;

determining the entity facts that are accessible to the member according to the entity fact access control lists;

determining, based on member information of the member, and entity facts that are accessible to the member, search result data including data describing entities and entity facts relevant to the query; and

providing search results, based on the search result data, to the member of the enterprise, the search results including data describing the entities and entity facts determined to be relevant to the query.

2. The method of claim 1, wherein member information of the member includes at least one of the member'"'"'s role in the enterprise, an organizational structure of the enterprise, the member'"'"'s contacts, and the member'"'"'s relationship with the provided query.

3. The method of claim 2, wherein the member'"'"'s relationship with the provided query includes:

determining, by a data system, that the query corresponds to at least one of an entity fact and entity of the enterprise;

relating, by the data system, the query with the at least one of an entity fact and entity; and

determining, by the data system, the member'"'"'s relationship with the at least one of the entity fact and entity.

4. The method of claim 1, further comprising:

ranking the search result data based on a relevance of the entities and entity facts in relation to the query and member information of the member; and

providing the search results, based on the search result data, in an order of the rankings to the member of the enterprise including data describing the entities and entity facts determined to be relevant to the query.

5. The method of claim 4, wherein the ranking the search result data based on the relevance of the entities and entity facts in relation to the query and member information, further comprises:

determining a quality score of the entities and entity facts, the quality score describing the relevance of the entities and entity facts in relation to the query and member information of the member; and

ranking the search result data based in part on the quality score of the entities and entity facts.

6. The method of claim 5, wherein the quality score is based on one or more parameters, the one or more parameters being at least one of a mean reciprocal rank, a mean precision, an average click position, a click through rate, and an abandonment rate.

7. The method of claim 4, further comprises:

receiving a selection from the search results provided to the member; and

adjusting a quality score of the entities and entity facts based on the selection from the search results provided to the member.

8. The method of claim 7, wherein the adjusting occurs by auto-tuning without member input.

9. A system, comprising:

a processor; and

a computer-readable medium coupled to the processor and having instructions stored thereon, which, when executed by the processor, cause the processor to perform operations comprising;

receiving a query from a member of an enterprise;

searching an index that includes member information of members of the enterprise and documents of the enterprise, wherein;

the documents include data describing entities and entity facts related to the enterprise and relationships among the entities, and each document has a respective access control list specifying access privileges to the document for members of the enterprise;

the entity facts are identified from the entities of the documents of the enterprise;

each entity fact describes at least one feature of the entity, wherein the feature of the entity is a relationship between the entity and another entity and wherein each entity fact is derived from one or more corresponding documents in which the entity fact is described; and

the index includes data defining access privileges to the data describing the entities and the entity facts according to respective entity fact access control lists, wherein each entity fact access control list is different from the access control lists provided for the documents of the enterprise, and each entity fact inherits an access control list of a document from which the entity fact is derived, wherein deriving the entity facts comprises selecting each document from the documents, and for the selected document;

determining a first entity identified within the document;

determining a second entity identified within the document;

determining a relationship between the first entity and the second entity that is described within the document; and

generating, as the entity fact, data describing the first entity, the second entity, and the relationship between the first entity and the second entity as described in the document;

wherein multiple entity facts are derived from a selected document;

determining the entity facts that are accessible to the member according to the entity fact access control lists;

determining, based on member information of the member, and entity facts that are accessible to the member, search result data including data describing entities and entity facts relevant to the query; and

providing search results, based on the search result data, to the member of the enterprise, the search results including data describing the entities and entity facts determined to be relevant to the query.

10. The system of claim 9, wherein member information of the member includes at least one of the member'"'"'s role in the enterprise, an organizational structure of the enterprise, the member'"'"'s contacts, and the member'"'"'s relationship with the provided query.

11. The system of claim 10, wherein the member'"'"'s relationship with the provided query includes:

determining that the query corresponds to at least one of an entity fact and entity of the enterprise;

relating the query with the at least one of an entity fact and entity; and

determining the member'"'"'s relationship with the at least one of the entity fact and entity.

12. The system of claim 9, further comprising:

ranking the search result data based on a relevance of the entities and entity facts in relation to the query and member information of the member; and

providing the search results, based on the search result data, in an order of the rankings to the member of the enterprise including data describing the entities and entity facts determined to be relevant to the query.

13. The system of claim 12, wherein the ranking the search result data based on the relevance of the entities and entity facts in relation to the query and member information, further comprises:

determining a quality score of the entities and entity facts, the quality score describing the relevance of the entities and entity facts in relation to the query and member information of the member; and

ranking the search result data based in part on the quality score of the entities and entity facts.

14. The system of claim 12, further comprises:

receiving a selection from the search results provided to the member; and

adjusting a quality score of the entities and entity facts based on the selection from the search results provided to the member.

15. The system of claim 14, wherein the adjusting occurs by auto-tuning without member input.

16. A computer-readable medium having instructions stored thereon, which, when executed by a processor, cause the processor to perform operations, comprising:

receiving a query from a member of an enterprise;

searching an index that includes member information of members of the enterprise and documents of the enterprise, wherein;

the documents include data describing entities and entity facts related to the enterprise and relationships among the entities, and each document has a respective access control list specifying access privileges to the document for members of the enterprise;

the entity facts are identified from the entities of the documents of the enterprise;

each entity fact describes at least one feature of the entity, wherein the feature of the entity is a relationship between the entity and another entity and wherein each entity fact is derived from one or more corresponding documents in which the entity fact is described; and

the index includes data defining access privileges to the data describing the entities and the entity facts according to respective entity fact access control lists, wherein each entity fact access control list is different from the access control lists provided for the documents of the enterprise, and each entity fact inherits an access control list of a document from which the entity fact is derived, wherein deriving the entity facts comprises selecting each document from the documents, and for the selected document;

determining a first entity identified within the document;

determining a second entity identified within the document;

determining a relationship between the first entity and the second entity that is described within the document; and

generating, as the entity fact, data describing the first entity, the second entity, and the relationship between the first entity and the second entity as described in the document;

wherein multiple entity facts are derived from a selected document;

determining the entity facts that are accessible to the member according to the entity fact access control lists;

determining, based on member information of the member, and entity facts that are accessible to the member, search result data including data describing entities and entity facts relevant to the query; and

providing search results, based on the search result data, to the member of the enterprise, the search results including data describing the entities and entity facts determined to be relevant to the query.

17. The computer-readable medium of claim 16, wherein member information of the member includes at least one of the member'"'"'s role in the enterprise, an organizational structure of the enterprise, the member'"'"'s contacts, and the member'"'"'s relationship with the provided query.

18. The computer-readable medium of claim 16, further comprising:

ranking the search result data based on a relevance of the entities and entity facts in relation to the query and member information of the member; and

providing the search results, based on the search result data, in an order of the rankings to the member of the enterprise including data describing the entities and entity facts determined to be relevant to the query.

19. The computer-readable medium of claim 18, wherein the ranking the search result data based on the relevance of the entities and entity facts in relation to the query and member information, further comprises:

determining a quality score of the entities and entity facts, the quality score describing the relevance of the entities and entity facts in relation to the query and member information of the member; and

ranking the search result data based in part on the quality score of the entities and entity facts.

20. The computer-readable medium of claim 18, further comprises:

receiving a selection from the search results provided to the member; and

adjusting a quality score of the entities and entity facts based on the selection from the search results provided to the member.