Method for performing sequence labelling on queries

• US 10,559,018 B2
• Filed: 10/19/2017
• Issued: 02/11/2020
• Est. Priority Date: 12/29/2014
• Status: Active Grant

• Current Assignee: eBay Inc.
• Original Assignee: eBay Inc.
• Inventors: Ruvini, Jean-David, Mohan, Sunil, Mukund, Smruthi, Sawaf, Hassan
• Primary Examiner(s): Stevens, Robert
• Application Number:
  US15/788,581
• Publication Number:
  US 20180040049A1
• Time in Patent Office: 845 Days
• Field of Search: 707732
• CPC Class Codes: G06F 16/24573 using data annotations, e.g. user-defined metadata
• CPC Class Codes: G06F 16/288 Entity relationship models
• CPC Class Codes: G06F 16/9535 Search customisation based on user profiles and personalisation
• CPC Class Codes: G06Q 30/0623 Item investigation

Abstract

A method of propagating annotations of content items to a search query is disclosed. A strength of a correspondence between a search query and a listing of an item on a network-based publication system is determined. The strength of the correspondence is based on an analysis of a set of actions by a set of users who submitted the search query. A set of annotations is generated. The set of annotations is propagated to a search engine and used to enhance search results.

Claims

1. A method comprising:

determining a strength of a correspondence between a search query and a data field corresponding to a listing of an item on a network-based publication system, the search query including a keyword, the determining of the strength of the correspondence based on an analysis of a set of actions previously taken by a set of users after submitting a set of previous search queries including the keyword;

generating a set of annotations corresponding to the search query, the generating including identifying a type of the item and a confidence measure pertaining to the identifying of the type of the item;

propagating the set of annotations to a search engine for use in enhancing search results generated by the search engine; and

providing the enhanced search results generated by the search engine for presentation, the enhanced search results including a filtering and/or ordering of the search results based on the type of the item and the confidence measure.

2. The method of claim 1, wherein the strength of the correspondence is based on an amount of revenue associated with the set of actions by the set of users who submitted the search query.

3. The method of claim 1, wherein the strength of the correspondence is based on rankings of types of the set of actions, the types of the set of actions including viewing the listing of the item, watching the listing of the item, bidding on the item, or purchasing the item.

4. The method of claim 1, wherein the generating of the set of annotations includes generating an annotation identifying a value of a property of the item.

5. The method of claim 1, wherein the enhancing of the search results includes reducing an expansion of the search results based on the strength of the correspondence transgressing a strength threshold.

6. The method of claim 1, wherein the enhancing of the search results includes increasing a relevancy of a search result item by an amount based on the set of annotations propagated to the search query.

7. The method of claim 6, further comprising determining the amount based on the strength of the correspondence.

8. A system comprising:

one or more modules implemented by one or more processors, the one or more modules configured to, at least;

determine a strength of a correspondence between a search query and a data field corresponding to a listing of an item on a network-based publication system, the search query including a keyword, the determining of the strength of the correspondence based on an analysis of a set of actions previously taken by a set of users after submitting a set of previous search queries including the keyword;

generate a set of annotations corresponding to the search query, the generating including identifying a type of the item and a confidence measure pertaining to the identifying of the type of the item;

propagate the set of annotations to a search engine for use in enhancing search results generated by the search engine; and

provide the enhanced search results generated by the search engine for presentation, the enhanced search results including a filtering and/or ordering of the search results based on the type of the item and the confidence measure.

9. The system of claim 8, wherein the strength of the correspondence is based on an amount of revenue associated with the set of actions by the set of users who submitted the search query.

10. The system of claim 8, wherein the strength of the correspondence is based on rankings of types of the set of actions, the types of the set of actions including viewing the listing of the item, watching the listing of the item, bidding on the item, or purchasing the item.

11. The system of claim 8, wherein the generating of the set of annotations includes generating an annotation identifying a value of a property of the item.

12. The system of claim 8, wherein the enhancing of the search results includes reducing an expansion of the search results based on the strength of the correspondence transgressing a strength threshold.

13. The system of claim 8, wherein the enhancing of the search results includes increasing a relevancy of a search result item by an amount based on the set of annotations propagated to the search query.

14. The system of claim 13, further comprising determining the amount based on the strength of the correspondence.

15. A non-transitory machine readable medium comprising a set of instructions that, when executed by a processor, causes the processor to perform operations, the operations comprising:

determining a strength of a correspondence between a search query and a data field corresponding to a listing of an item on a network-based publication system, the search query including a keyword, the determining of the strength of the correspondence based on an analysis of a set of actions previously taken by a set of users after submitting a set of previous search queries including the keyword;

generating a set of annotations corresponding to the search query, the generating including identifying a type of the item and a confidence measure pertaining to the identifying of the type of the item;

propagating the set of annotations to a search engine for use in enhancing search results generated by the search engine; and

providing the enhanced search results generated by the search engine for presentation, the enhanced search results including a filtering and/or ordering of the search results based on the type of the item and the confidence measure.

16. The non-transitory machine readable medium of claim 15, wherein the strength of the correspondence is based on an amount of revenue associated with the set of actions by the set of users who submitted the search query.

17. The non-transitory machine readable medium of claim 15, wherein the strength of the correspondence is based on rankings of types of the set of actions, the types of the set of actions including viewing the listing of the item, watching the listing of the item, bidding on the item, or purchasing the item.

18. The non-transitory machine readable medium of claim 15, wherein the generating of the set of annotations corresponding to the title includes generating an annotation identifying a value of a property of the item.

19. The non-transitory machine readable medium of claim 15, wherein the enhancing of the search results includes reducing an expansion of the search results based on the strength of the correspondence transgressing a strength threshold.

20. The non-transitory machine readable medium of claim 15, wherein the enhancing of the search results includes increasing a relevancy of a search result item by an amount based on the set of annotations propagated to the search query.