Presenting search results according to query domains

• US 9,684,741 B2
• Filed: 06/05/2009
• Issued: 06/20/2017
• Est. Priority Date: 06/05/2009
• Status: Active Grant

• Current Assignee: Microsoft Technology Licensing LLC (Microsoft Corporation)
• Original Assignee: Microsoft Technology Licensing LLC (Microsoft Corporation)
• Inventors: Acero, Alejandro, Zweig, Geoffrey, Nguyen, Patrick, Li, Xiao
• Primary Examiner(s): TORGRIMSON, TYLER J
• Application Number:
  US12/479,371
• Publication Number:
  US 20100312782A1
• Time in Patent Office: 2,937 Days
• Field of Search: None
• CPC Class Codes: G06F 16/9038 Presentation of query results
• CPC Class Codes: G06N 20/00 Machine learning
• CPC Class Codes: G10L 15/26 Speech to text systems G10L15/08 takes precedence
• CPC Class Codes: G10L 2015/025 Phonemes, fenemes or fenones being the recognition units

Abstract

A query may be applied against search engines that respectively return a set of search results relating to various items discovered in the searched data sets. However, presenting numerous and varied search results may be difficult on mobile devices with small displays and limited computational resources. Instead, search results may be associated with search domains representing various information types (e.g., contacts, public figures, places, projects, movies, music, and books) and presented by grouping search results with associated query domains, e.g., in a tabbed user interface. The query may be received through an input device associated with a particular input domain, and may be transitioned to the query domain of a particular search engine (e.g., by recognizing phonemes of a voice query using an acoustic model; matching phonemes with query terms according to a pronunciation model; and generating a recognition result according to a vocabulary of an n-gram language model.)

Claims

1. A method of presenting search results in response to a query on behalf of a user of a device having a processor and a display, the method comprising:

executing on the processor instructions configured to;

execute the query on at least one search engine to generate a plurality of search results;

for respective search results generated by the at least one search engine;

compute a search result relevance score for the search result;

identify at least one topic semantically associated with the search result,

identify a topic type of the at least one topic;

identify at least one distinctive characteristic of the search result; and

associating the search result with the identified topic type according to the at least one identified distinctive characteristic by evaluating the search result with a classifier configured to identify the topic types of a search result having the at least one distinctive characteristic;

for respective topic types, compute a topic type relevance score of the topic type based on the relevance scores of the search results associated with the topic type; and

present on the display;

a plurality of topic type groups respectively representing a topic type and comprising a plurality of search results semantically associated with at least one topic of the topic type, the topic type groups sorted in descending order by the relevance scores of the search results within the topic type group, the groups comprising tabs in a tabbed user interface; and

within respective topic type groups, the search results semantically associated with at least one topic of the topic type sorted in descending order by the search result relevance scores.

2. The method of claim 1:

the device comprising at least one input component associated with an input domain, and

the instructions configured to receive the query through at least one input component.

3. The method of claim 2, receiving the query comprising:

presenting on the display an input control associated with an input component and configured to receive the query through the input component; and

upon detecting a user activation of the input control, receiving the query through the input component associated with the input control.

4. The method of claim 2, the instructions configured to, after receiving the query and before executing the query on the at least one search engine, normalize the query.

5. The method of claim 4:

at least one input component of the device associated with an input domain having an input grammar; and

the instructions comprising;

after receiving the query, normalizing the query according to an input grammar associated with the input component receiving the query.

6. The method of claim 4:

the device comprising a query term normalization database storing at least one common query term; and

normalizing the query comprising;

identifying at least one common query term in the query term normalization database that approximates a query term of the textual query, and

upon identifying the at least one query term, replacing the query term in the textual query with the common query term.

7. The method of claim 2:

at least one input component of the device comprising an audio input component;

the query comprising an audio query received through an audio input component;

at least one search engine comprising a textual search engine configured to handle textual queries; and

the instructions configured to, after receiving the query through the audio input component and before executing the query on the textual search engine, transitioning the audio query to a textual query.

8. The method of claim 7, transitioning the query comprising:

identifying at least one phoneme in the query according to an acoustic model;

identifying at least one query term corresponding to the at least one phoneme according to a pronunciation model; and

generating a recognition result from the at least one query term according to a vocabulary of an n-gram language model.

9. The method of claim 7, the instructions configured to, after recognizing the audio query to a textual query and before executing the query on the textual search engine, confirm the textual query with the user.

10. The method of claim 1:

respective topic types associated with search results identifiable by at least one distinctive characteristic, and

associating the search results with topic types comprising;

identifying at least one distinctive characteristic of the search result, and

associating the search result with the at least one topic type according to the at least one identified distinctive characteristic.

11. The method of claim 10, associating the search result with a topic types comprising:

evaluating the search result with a statistical classifier configured to identify the topic types of a search result having the at least one distinctive characteristic.

12. The method of claim 1, presenting the topic types comprising:

present on the display a grouped interface comprising groups allocated to respective topic types and presenting a name of the topic type; and

upon receiving a user selection of a topic type group, present on the display;

the topic type associated with the topic type group, and

the search results associated with the topic type.

13. The method of claim 12, the groups comprising tabs in a tabbed user interface.

14. The method of claim 1:

the instructions configured to identify at least two search results associated with a topic type that is associated with a topic, and

presenting the topic type groups comprising;

presenting on the display the at least two search results grouped together within the topic type.

15. The method of claim 1:

the instructions defining a minimally sufficient relevance score; and

presenting the topic type groups comprising;

presenting on the display the topic types grouped with associated search results having a relevance score at least equaling the minimally sufficient relevance score.

16. The method of claim 15:

presenting the topic type groups comprising;

presenting on the display the topic type groups associated with at least one search result having a topic type relevance score at least equaling the minimally sufficient topic type relevance score.

17. A system configured to present search results in response to a query on behalf of a user of a device having a display and a memory, the system comprising:

a query executor comprising instructions stored in the memory that, when executed on the processor, cause the device to execute the query on at least one search engine to generate one a plurality of search results;

a topic type associator comprising instructions stored in the memory that, when executed on the processor, cause the device to, for respective search results generated by the at least one search engine;

compute a search result relevance score for the search result;

identify at least one topic semantically associated with the search result;

identify a topic type of the at least one topic;

identify at least one distinctive characteristic of the search result; and

associate the search result with the identified topic type according to the at least one identified distinctive characteristic by evaluating the search result with a classifier configured to identify the topic types of a search result having the at least one distinctive characteristic;

for respective topic types, compute a topic type relevance score of the topic type based on the relevance score of the search results associated with the topic type; and

a search result presenter comprising instructions stored in the memory that, when executed on the processor, cause the device to present on the display;

a plurality of topic type groups respectively representing a topic type and comprising a plurality of search results semantically associated with at least one topic of the topic type, the groups sorted in descending order by the relevance scores of the search results associated with the topic type group, the groups comprising tabs in a tabbed user interface; and

within respective topic type groups, the search results semantically associated with at least one topic of the topic type sorted in descending order by the search result relevance scores.