DIALOG-ENHANCED CONTEXTUAL SEARCH QUERY ANALYSIS

• US 20120323948A1
• Filed: 06/16/2011
• Published: 12/20/2012
• Est. Priority Date: 06/16/2011
• Status: Active Grant

• Current Assignee: Microsoft Technology Licensing LLC (Microsoft Corporation)
• Original Assignee: Microsoft Corporation
• Inventors: LI, SHIPENG, WANG, YE-YI, NI, XIAOCHUAN, SUN, JIAN-TAO, CHEN, ZHENG, ZHANG, ZHIMIN, LI, XIAOLONG, HUANG, XUEDONG, JIANG, LI
• Granted Patent: US 9,336,298 B2
• Time in Patent Office: Days
• US Class Current: 707/765
• CPC Class Codes: G06F 16/3323 using document space presentation or visualization, e.g. category, hierarchy or range presentation and selection

Abstract

Embodiments of the present invention relate to systems, methods, and computer-storage media for a method of contextually analyzing terms within a search query. In one embodiment, a received search query is classified into a domain category. Additionally, information is assigned to a schema associated with the domain by analyzing the search query. Further, at least one search result that helps a user complete a task within the domain is provided based on the information in the schema.

Claims

1. One or more computer-storage media having computer-executable instructions embodied thereon that, when executed, perform a method of analyzing terms within a search query using a domain specific analysis, the method comprising:

receiving a search query that initiates a search session;

classifying the search query into a subject-matter domain;

determining a task within the subject-matter domain that a user intends to complete through the search session;

filling slots in a dialog schema with initial information derived from terms in the search query, wherein the dialog schema comprises slots for information that is used to complete the task;

determining that one or more slots in the dialog schema are not associated with any information;

providing a dialog feature that invites the user to provide additional information used to complete the task;

receiving the additional information; and

providing at least one search result based on the initial information and the additional information in the dialog schema, wherein the at least one search result is related to completing the task.

2. The media of claim 1, wherein each slot is derived through an analysis of information provided by users conducting searches within the subject-matter domain previously, wherein the analysis compares n-grams in the search query with n-grams derived from entities in a domain knowledge base, wherein a match with an infrequently occurring n-gram indicates a possible match between an entity in which the n-gram is present.

3. The media of claim 1, wherein each slot contains information that is consumable by a search application to identify search results the user is seeking.

4. The media of claim 1, wherein the at least one search result comprises an application that helps the user complete the task.

5. The media of claim 4, wherein the application comprises fields that are prefilled with the initial information and the additional information.

6. The media of claim 1, wherein the method further comprises:

receiving one or more additional search queries from the user;

determining that the one or more additional search queries are part of the search session; and

assigning further information to empty slots in the dialog schema using terms from the one or more additional search queries.

7. The media of claim 1, wherein the dialog feature asks for the additional information in an order in which people typically provide information, wherein the order is determined by performing a historical analysis of user behavior associated with the task.

8. One or more computer-storage media having computer-executable instructions embodied thereon that, when executed by a computing device, perform a method of analyzing terms within a search query using a domain specific analysis, the method comprising:

receiving a search query;

classifying the search query into a subject-matter domain;

classifying the search query into a task within the subject-matter domain using a machine classifier that is trained to classify a query into one or more tasks that are within the subject-matter domain; and

providing at least one search result related to completing the task.

9. The media of claim 8, wherein the method further comprises filling slots in a dialog schema with initial information derived from terms in the search query, wherein the dialog schema comprises slots for information that is used to complete the task.

10. The media of claim 9, wherein the slots are assigned information based on placement of the terms within the search query.

11. The media of claim 9, further comprising:

determining that one or more slots in the dialog schema are not associated with information.

12. The media of claim 11, further comprising:

providing a user interface with a dialog feature that invites a user to provide additional information used to complete the task; and

receiving the additional information,

wherein the at least one search result is based on the additional information.

13. The media of claim 9, wherein the at least one search result is an application used to complete the task, and wherein the method further comprises communicating information extracted from the search query to the application.

14. The media of claim 9, wherein the method further comprises communicating an interface that asks for confirmation that a user intends to complete the task and suggests one or more additional tasks that have a confidence factor above a threshold that the user can select as an alternative.

15. A method analyzing terms within a search query using a domain specific analysis, the method comprising:

receiving a first search query comprising a first plurality of search terms;

classifying the first search query into a domain;

assigning a first information to a first slot associated with a domain schema, wherein the first information is based on the first plurality of search terms;

receiving a second search query comprising a second plurality of search terms;

classifying the second search query into the domain, thereby determining the first search query and the second search query are part of a single search session;

assigning a second information to a second slot associated with the domain schema, wherein the second information is derived from the second plurality of search terms; and

providing search results based on the first information and the second information.

16. The method of claim 15, wherein the first and second slots are within a dialog schema that comprises a set of slots.

17. The method of claim 16, wherein the method further comprises:

determining that a third slot within the dialog schema is empty;

providing a dialog feature that requests information that can be used to fill the third slot;

receiving a response to the dialog feature; and

assigning third information to the third slot, wherein the second information is derived from the response, and

wherein the search results are also based on the third information.

18. The method of claim 16, wherein the search results are not provided until after a threshold number of slots of the dialog schema are filled.

19. The method of claim 15, wherein the domain is determined using a statistical classifier.

20. The method of claim 15, wherein the first information is generated by comparing the first plurality of search terms with a knowledge base to disambiguate a term.