Entity detection and extraction for entity cards

• US 9,158,846 B2
• Filed: 06/10/2010
• Issued: 10/13/2015
• Est. Priority Date: 06/10/2010
• Status: Active Grant

• Current Assignee: Microsoft Technology Licensing LLC (Microsoft Corporation)
• Original Assignee: Microsoft Technology Licensing LLC (Microsoft Corporation)
• Inventors: Radlinski, Filip, Craswell, Nick, Shokouhi, Milad, Ahari, Sanaz, Agrawal, Nitin, Zhou, Song, Awan, Muhammad Aatif, Billerbeck, Bodo, Hoad, Timothy
• Primary Examiner(s): Reyes, Mariela
• Assistant Examiner(s): Almani, Mohsen
• Application Number:
  US12/813,390
• Publication Number:
  US 20110307483A1
• Time in Patent Office: 1,951 Days
• Field of Search: 707/999.003, 707/1, 707/748, 707/731
• US Class Current: 1/1
• CPC Class Codes: G06F 16/2455 Query execution
• CPC Class Codes: G06F 16/24578 using ranking
• CPC Class Codes: G06F 16/248 Presentation of query results
• CPC Class Codes: G06F 16/951 Indexing; Web crawling techniques
• CPC Class Codes: G06F 16/9535 Search customisation based on user profiles and personalisation
• CPC Class Codes: G06F 16/9538 Presentation of query results

Abstract

A system and method are provided for detecting entity information contained within search results. The detected entity information can be used to determine a category of entity as well as a specific entity within the search results. Entity information can be extracted from the documents associated with the search results. This information can be used as part of the information for an entity card, which can be displayed to a user in conjunction with and/or in place of the search results.

Claims

1. A system for determining an entity associated with a search query, the system comprising:

one or more processors; and

one or more computer-storage media storing computer-useable instructions that, when executed by the one or more processors, cause the one or more processors to perform a method comprising;

obtaining a plurality of results responsive to a search query;

matching responsive results of the plurality of results to a plurality of category templates corresponding to category-oriented sites, a category template comprising a presentation format for a corresponding category-oriented site, a category-oriented site being defined as a site that tracks a status of a plurality of entities in an entity category;

assigning an entity category, selected from a plurality of entity categories, to the search query based on the responsive results matching the plurality of category templates, the assigning comprising generating category scores for the plurality of entity categories based on an amount of the responsive results that match a particular one of the plurality of category templates, the entity category being assigned based on having the highest category score of the plurality of entity categories;

identifying an entity based on the entity belonging to the assigned entity category and at least one or more data fields extracted from the matched responsive results, the one or more data fields being extracted using extraction format components of the plurality of category templates, the one or more data fields being different from a URL component;

retrieving stored entity information regarding the identified entity;

incorporating the retrieved entity information and at least one responsive result into an entity card;

displaying the responsive results in a search result listing, the search result listing not including the at least one incorporated responsive result; and

displaying the entity card.

2. The system of claim 1, wherein in the assigning the entity category additional weight is given to a category score based on the amount of the responsive results that match the particular one of the plurality of category templates.

3. The system of claim 1, wherein the matching comprises matching a portion of a uniform resource locator (URL) for a document of the responsive results with a URL component of a category template of the plurality of category templates.

4. The system of claim 1, the method further comprising supplementing the stored entity information with additional entity information extracted from the responsive results.

5. The system of claim 2, wherein the plurality of entity categories comprise a movie category, a game category, a restaurant category, and a consumer goods category.

6. The system of claim 1, wherein identifying the entity comprises:

generating an entity score for each one of a plurality of entities based on at least the one or more data fields extracted from the responsive results; and

identifying the entity having a highest entity score.

7. One or more computer-storage media hardware devices storing computer-useable instructions that, when executed by a computing device, perform a method for determining an entity associated with a search query, comprising:

obtaining a plurality of results responsive to a search query;

matching one or more responsive results of the plurality of results to at least one category template corresponding to a category-oriented site, a category-oriented site being defined as a site that tracks a status of a plurality of entities in an entity category;

assigning an entity category to the search query based on the one or more responsive results matching the at least one category template, the assigning comprising generating category scores for a plurality of entity categories based on an amount of the one or more responsive results that match a particular one of the plurality of category templates, the entity category being assigned based on having the highest category score of the plurality of entity categories;

identifying a plurality of entities corresponding to the assigned entity category in the plurality of responsive results, each identified entity identified based on belonging to the entity category;

extracting secondary identification information corresponding to each identified entity from the plurality of responsive results;

sorting the identified plurality of entities into a plurality of entity groups based on the extracted secondary identification information;

selecting an entity group as corresponding to the search query;

determining one or more responsive results corresponding to the selected entity group;

incorporating the extracted secondary identification information for the selected entity group into an entity card; and

displaying the entity card and a portion of the responsive results, the portion of the responsive results excluding the determined one or more responsive results corresponding to the entity of the selected entity group.

8. The one or more computer-storage media hardware devices of claim 7, wherein determining one or more responsive results corresponding to the entity of the selected entity group comprises identifying responsive results from which secondary identification information for the selected entity group was extracted.

9. The one or more computer-storage media hardware devices of claim 7, wherein matching one or more responsive results to at least one category template corresponding to category-oriented sites comprises matching a portion of a uniform resource locator (URL) for a document of the responsive results with a URL component of a category template.

10. The one or more computer-storage media hardware devices of claim 7, further comprising:

identifying, for the selected entity group, at least one secondary identification information field extracted from the responsive results that have conflicting extracted values;

assigning a stored value to the information field having the conflicting extracted values in the extracted entity information.

11. The one or more computer-storage media hardware devices of claim 7, wherein selecting an entity group comprises selecting an entity group containing extracted secondary information from a highest number of responsive results.

12. The one or more computer-storage media hardware devices of claim 7, wherein selecting an entity group comprises selecting an entity group containing extracted secondary information from at least a plurality of documents.

13. The one or more computer-storage media hardware devices of claim 7, wherein extracting secondary information from the one or more responsive results comprises extracting data fields from one or more documents based on an open form category template.

14. A method for determining an entity associated with a search query, comprising:

obtaining results responsive to a search query;

matching documents from the results to at least one category template corresponding to a category-oriented site, the at least one category template comprising a presentation format for the corresponding category-oriented site, a category-oriented site being defined as a site that tracks a status of a plurality of entities in an entity category;

generating category scores for a plurality of entity categories based on the results, wherein additional weight is given to a category score based on an amount of a plurality of the documents that match the at least one category template;

assigning one or more entity categories of the plurality of entity categories to the search query based on the category scores;

identifying a plurality of entities based on at least the assigned one or more entity categories and at least one or more data fields extracted from the one or more matched documents, the one or more data fields being extracted using extraction format components of the at least one category template, the one or more data fields being different from a URL component;

extracting entity information corresponding to the plurality of identified entities from the plurality of responsive results;

incorporating the extracted entity information for each of the plurality of identified entities into a plurality of entity cards based on the search query; and

displaying the plurality of entity cards.

15. The method of claim 14, further comprising excluding from display at least one non-matched responsive result.

16. The method of claim 15, wherein excluding from display the at least one non-matched responsive result comprises providing a condensed representation for the at least one non-matched responsive result, the condensed representation requiring at least one additional user action to display the at least one non-matched responsive result.

17. The method of claim 16, wherein the at least one additional user action required to display the at least one non-matched responsive result comprises clicking on a displayed object.

18. The method of claim 16, wherein the at least one additional user action required to display the at least one non-matched responsive result comprises selecting the non-matched responsive result from a drop-down menu.

19. The method of claim 14, wherein extracting information from the one or more matched documents comprises extracting data fields from the one or more matched documents based on an open form category template.

20. The method of claim 14, wherein the plurality of identified entities are from the same entity category.