Method and system for semantic search against a document collection

• US 9,710,518 B2
• Filed: 07/17/2015
• Issued: 07/18/2017
• Est. Priority Date: 07/13/2011
• Status: Active Grant

• Current Assignee: Microsoft Technology Licensing LLC (Microsoft Corporation)
• Original Assignee: LinkedIn Corporation (Microsoft Corporation)
• Inventors: Tunkelang, Daniel, Cheng, Heyning
• Primary Examiner(s): Morrison, Jay
• Application Number:
  US14/802,466
• Publication Number:
  US 20150324431A1
• Time in Patent Office: 732 Days
• Field of Search: None
• CPC Class Codes: G06F 16/00 Information retrieval; Database structures therefor; File system structures therefor
• CPC Class Codes: G06F 16/242 Query formulation
• CPC Class Codes: G06F 16/243 Natural language query formulation
• CPC Class Codes: G06F 16/245 Query processing
• CPC Class Codes: G06F 16/24575 using context
• CPC Class Codes: G06F 16/284 Relational databases
• CPC Class Codes: G06F 16/9535 Search customisation based on user profiles and personalisation
• CPC Class Codes: G06F 7/00 Methods or arrangements for processing data by operating upon the order or content of the data handled logic circuits H03K19/00
• CPC Class Codes: G06Q 50/01 Social networking

Abstract

Disclosed in one example is a method for searching. In some examples, the method includes receiving an unstructured search query, parsing the unstructured search query into a plurality of structured search attributes using a search term attribute dictionary, performing a search of a structured database based upon the plurality of structured search attributes to identify a plurality of search results, populating a plurality of form fields of a search form with the plurality of structured search attributes, and displaying the search form in association with the search results.

Claims

1. A method comprising:

receiving a raw search field value;

determining a standardized search field from a plurality of standardized search fields for the raw search field value;

mapping the raw search field value to a standard field value, the standard field value corresponding to the standardized search field;

determining a plurality of similarity scores between the standard field value and a plurality of other standard field values based upon a cosine similarity using field values associated with member profiles;

using the plurality of similarity scores, selecting a second standard field value;

performing a search using the raw search field value, the standard field value, and the second standard field value to identify a plurality of search results;

displaying the plurality of search results.

2. The method of claim 1, wherein selecting the second standard field value includes determining that a similarity score between the standard field value the second standard field value exceeds a threshold.

3. The method of claim 1, wherein selecting the second standard field value includes determining that a similarity score between the standard field value the second standard field value exceeds all other similarity scores in the plurality of similarity scores.

4. The method of claim 1, wherein performing the search includes performing the search on a structured database.

5. The method of claim 4, wherein the structured database includes the plurality of standardized search fields, and performing the search on the structured database includes performing the search on the standardized search field.

6. The method of claim 1, further comprising:

receiving a user edit to the search, the user edit removing at least one of the standard field value, the second standard field value, and raw search field value; and

performing a new search based on the user edit.

7. The method of claim 1, further comprising populating a search form with the standard field value and the second standard field value in a representation of the standardized search field.

8. The method of claim 7, wherein displaying the plurality of search results includes displaying the search form.

9. The method of claim 7, wherein populating the search form includes populating the search form with the raw search field value.

10. A machine readable medium not having any transitory signals and comprising instructions that, when executed by at least one processor of a machine, cause the machine to perform operations to:

receive a raw search field value;

determine a standardized search field from a plurality of standardized search fields for the raw search field value;

map the raw search field value to a standard field value, the standard field value corresponding to the standardized search field;

determine a plurality of similarity scores between the standard field value and a plurality of other standard field values based upon a cosine similarity using field values associated with member profiles;

using the plurality of similarity scores, select a second standard field value;

perform a search using the raw search field value, the standard field value, and the second standard field value to identify a plurality of search results;

display the plurality of search results.

11. The machine readable medium of claim 10, wherein operations to select the second standard field value include operations to determine that a similarity score between the standard field value the second standard field value exceeds a threshold.

12. The machine readable medium of claim 10, wherein operations to select the second standard field value include operations to determine that a similarity score between the standard field value the second standard field value exceeds all other similarity scores in the plurality of similarity scores.

13. The machine readable medium of claim 10, further comprising operations to populate a search form with the standard field value and the second standard field value in a representation of the standardized search field.

14. The machine readable medium of claim 13, wherein operation to display the plurality of search results include operations to display the search form.

15. The machine readable medium of claim 13, wherein operations to populate the search form include operations to populate the search form with the raw search field value.

16. A system comprising:

one or more processors coupled to a memory device, the memory device containing instructions that, when executed by the one or more processors, cause the system to;

receive a raw search field value;

determine a standardized search field from a plurality of standardized search fields for the raw search field value;

map the raw search field value to a standard field value, the standard field value corresponding to the standardized search field;

determine a plurality of similarity scores between the standard field value and a plurality of other standard field values based upon a cosine similarity using field values associated with member profiles;

using the plurality of similarity scores, select a second standard field value;

perform a search using the raw search field value, the standard field value, and the second standard field value to identify a plurality of search results;

display the plurality of search results.

17. The system of claim 16, wherein operations to select the second standard field value include operations to determine that a similarity score between the standard field value the second standard field value exceeds a threshold.

18. The system of claim 16, wherein operations to select the second standard field value include operations to determine that a similarity score between the standard field value the second standard field value exceeds all other similarity scores in the plurality of similarity scores.

19. The system of claim 16, wherein operations to perform the search include operations to perform the search on the standardized search field in a structured database, the structured database including the plurality of standardized search fields.

20. The system of claim 19, wherein the instructions further cause the system to:

receive a user edit to the search, the user edit removing at least one of the standard field value, the second standard field value, and raw search field value; and

perform a new search based on the user edit.