Learning and using contextual content retrieval rules for query disambiguation

• US 10,579,652 B2
• Filed: 06/17/2014
• Issued: 03/03/2020
• Est. Priority Date: 06/17/2014
• Status: Active Grant

• Current Assignee: Microsoft Technology Licensing LLC (Microsoft Corporation)
• Original Assignee: Microsoft Technology Licensing LLC (Microsoft Corporation)
• Inventors: Bennett, Paul, Collins-Thompson, Kevyn, Sarkizova, Siranush, Shokouhi, Milad, Sloan, Marc
• Primary Examiner(s): Lu, Charles E
• Application Number:
  US14/306,500
• Publication Number:
  US 20150363485A1
• Time in Patent Office: 2,086 Days
• CPC Class Codes: G06F 16/24564 Applying rules; Deductive queries
• CPC Class Codes: G06F 16/332 Query formulation
• CPC Class Codes: G06F 16/3325 Reformulation based on results of preceding query
• CPC Class Codes: G06F 16/3331 Query processing
• CPC Class Codes: G06F 16/3338 Query expansion
• CPC Class Codes: G06F 16/86 Mapping to a database
• CPC Class Codes: G06F 16/951 Indexing; Web crawling techniques
• CPC Class Codes: G06F 16/9535 Search customisation based on user profiles and personalisation

Abstract

Various technologies related to generating and applying content retrieval rules are described herein. A content retrieval rule maps a combination of a query and a context to one of a query reformulation or content. The content retrieval rule is learned from search logs of a search engine, and is applied when the query having the context is received at the search engine.

Claims

1. A computing system comprising:

a processor; and

memory that comprises instructions that, when executed by the processor, cause the processor to perform acts comprising;

during a search session with a search engine, receiving a query from a client computing device that is in network communication with the computing system;

responsive to receiving the query, assigning a topic to the query from amongst a plurality of predefined topics, wherein the topic is being searched for by a user of the client computing device during the search session, wherein the topic is assigned based upon a previous query received from the client computing device during the search session, wherein the query and the previous query are labeled as belonging to the search session based upon the query being received within a threshold amount of time from when the previous query was received;

responsive to assigning the topic to the query, identifying a content retrieval rule in a database of content retrieval rules based upon a combination of the query and the topic assigned to the query, wherein the content retrieval rule maps the combination of the query and the topic to a query reformulation;

responsive to identifying the content retrieval rule, performing a search over a document collection using the query reformulation to acquire a ranked list of search results; and

transmitting the ranked list of search results to the client computing device, wherein the client computing device displays the ranked list of search results on a display thereof responsive to receiving the ranked list of search results.

2. The computing system of claim 1, wherein the search is performed by the search engine using both the query and the query reformulation, wherein the ranked list of search results is included in a search engine results page, and further wherein the ranked list of search results comprises content retrieved by the search engine based upon the query and the query reformulation.

3. The computing system of claim 2, the ranked list of search results comprises a first set of search results that is based upon the query and a second set of search results that is based upon the query reformulation.

4. The computing system of claim 2, the search engine results page comprising at least one of a map, an entity card, or an advertisement that is retrieved based upon the query reformulation.

5. The computing system of claim 1, wherein the topic is assigned to the query based upon a plurality of queries previously issued by a user who issued the query and the previous query, wherein each query in the plurality of queries is labeled as belonging to the search session.

6. The computing system of claim 5, wherein the topic is assigned to the query further based upon at least one of an interaction of the user with a social networking application or a network address of a computing device of the user.

7. The computing system of claim 1, the acts further comprising generating a plurality of content retrieval rules based upon search logs of a search engine, wherein the plurality of content retrieval rules comprise the content retrieval rule.

8. The computing system of claim 7, the acts further comprising assigning respective scores to the plurality of content retrieval rules, where the content retrieval rule is assigned to the query based upon a score assigned to the content retrieval rule.

9. A method comprising:

during a search session of a user with a search engine, receiving a query from a client computing device by way of a network connection, wherein the search session comprises a previous query set forth to the search engine by the user, wherein the query and the previous query are labeled as belonging to the search session based upon the query being issued to the search engine by the user within a threshold amount of time from when the previous query was issued to the search engine by the user;

assigning a topic to the query based upon the previous query set forth by the user during the search session, wherein the topic is assigned from amongst a plurality of predefined topics, and further wherein the topic is being searched for by the user during the search session;

identifying a content retrieval rule that is mapped to a combination of the query and the topic assigned to the query, the content retrieval rule maps the combination of the query and the topic to a query reformulation;

responsive to identifying the content retrieval rule, performing a search over a document collection using the query reformulation to acquire a ranked list of search results; and

transmitting the ranked list of search results to the client computing device, wherein the client computing device displays the ranked list of search results on a display thereof responsive to receiving the ranked list of search results.

10. The method of claim 9, further comprising:

performing another search over the document collection using the query.

11. The method of claim 10, further comprising:

outputting a search engine results page based upon the search and the another search, the search engine results page includes the ranked list of search results, the ranked list of search results includes a search result retrieved in the search and another search result retrieved in the another search.

12. The method of claim 11, further comprising:

highlighting a keyword of the reformulated query in the another search result, the query free of the highlighted keyword.

13. The method of claim 9, further comprising:

performing a second search over the document collection based upon the query;

generating a search engine results page based upon the query; and

presenting the query reformulation on the search engine results page as a suggested query, wherein the search over the document collection based upon the query reformulation is performed only after the query reformulation has been selected.

14. The method of claim 9, wherein the topic is assigned to the query based further upon demographics of the issuer of the query.

15. The method of claim 9, wherein the topic is assigned to the query based further upon search results returned by the search engine that were selected or not selected by the issuer of the query when the previous query was issued.

16. A computer-readable storage medium comprising instructions that, when executed by a processor, cause the processor to perform acts comprising:

while a user of a client computing device is participating in a search session with a search engine, receiving a query from the client computing device;

responsive to receiving the query, assigning a topic to the query from amongst a plurality of predefined topics, wherein the topic is assigned to the query based upon a previous query received from the client computing device during the search session, wherein the previous query and the query are labeled as belonging to the search session based upon the query being received within a threshold amount of time from when the previous query was received, and further wherein the topic is being searched for by the user of the client computing device during the search session;

responsive to assigning the topic to the query, identifying a content retrieval rule that is mapped to a combination of the query and the topic assigned to the query, the content retrieval rule specifies a query reformulation for the query;

performing a search over a document collection based upon the query reformulation, wherein performing the search results in generation of a ranked list of search results; and

transmitting the ranked list of search results to the client computing device, wherein the client computing device displays the ranked list of search results on a display thereof in response to receiving the ranked list of search results.

17. The computer-readable storage medium of claim 16, wherein the plurality of predefined topics are based upon topics assigned to web pages in search logs of the search engine.

18. The computer-readable storage medium of claim 16, wherein the topic assigned to the query is further based upon demographics of the user who issued the query.

19. The computer-readable storage medium of claim 16, wherein the topic is assigned to the query based upon topics assigned to other queries labeled as belonging to the search session.

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

performing a second search over the document collection based upon the query, wherein the ranked list of search results comprises at least one search result identified in the second search.