Automatic execution of actions responsive to search queries

• US 10,769,371 B1
• Filed: 11/28/2017
• Issued: 09/08/2020
• Est. Priority Date: 11/28/2017
• Status: Active Grant

• Current Assignee: Amazon Technologies, Inc. (Amazon.com, Inc.)
• Original Assignee: Amazon Technologies, Inc. (Amazon.com, Inc.)
• Inventors: Barrientos, Ivan, Bulusu, Prakash, Mishra, Pragyana K.
• Primary Examiner(s): Nguyen, Kim T
• Application Number:
  US15/824,477
• Time in Patent Office: 1,015 Days
• Field of Search: None
• CPC Class Codes: G06F 16/243 Natural language query formulation
• CPC Class Codes: G06F 16/9535 Search customisation based on user profiles and personalisation
• CPC Class Codes: G06F 3/0482 Interaction with lists of selectable items, e.g. menus
• CPC Class Codes: G06F 40/247 Thesauruses; Synonyms
• CPC Class Codes: G06F 40/274 Converting codes to words; Guess-ahead of partial word inputs
• CPC Class Codes: G06F 40/49 using very large corpora, e.g. the web

Abstract

Systems, methods, and computer-readable media are disclosed for systems and methods for automatic execution of actions responsive to queries. Example methods include determining a set of seed words for input at a first search engine, determining a first autocomplete result associated with a first word at the first search engine, and determining a second autocomplete result associated with the first autocomplete result at the first search engine. Example methods include determining a set of user interaction histories that comprises queries input at a second search engine, and determining that a first user interaction history includes a query with a portion of the second autocomplete result. Example methods include determining a user action performed after the query, and associating the user action with the query in a database, such that subsequent inputs of the query cause the first user action to automatically occur.

Claims

1. A method comprising:

determining, by one or more computer processors coupled to at least one memory, a first set of verbs that represent digital user actions;

determining a set of synonyms for verbs in the first set of verbs;

selecting a second set of verbs from the set of synonyms;

determining a set of seed verbs comprising the first set of verbs and the second set of verbs;

determining a prompt phrase using a first seed verb of the set of seed verbs, wherein the prompt phrase is a question comprising the first seed verb;

sending the prompt phrase to a first search engine for input as a first search query;

determining a first autocomplete result from the first search engine responsive to input of the first search query;

determining that the first autocomplete result comprises the first seed verb;

sending the first autocomplete result to the first search engine for input as a second search query;

determining a second autocomplete result from the first search engine responsive to input of the second search query;

determining a set of autocompleted results comprising the first autocomplete result and the second autocomplete result;

extracting the first seed verb and a first noun from the first autocomplete result;

determining a search history associated with a user account for a second search engine;

determining that the search history comprises a fourth search query, the fourth search query comprising the first seed verb and the first noun;

determining first browsing data associated with the user account, wherein the first browsing data is for a first time interval preceding the fourth search query;

determining second browsing data associated with the user account, wherein the second browsing data is for a second time interval after the fourth search query;

determining that the second browsing data comprises a user action of changing a device setting;

associating the fourth search query with the user action of changing the device setting in a user intended action database;

determining that the fourth search query is input at the second search engine at a user device; and

causing a device setting control user interface to be presented at the user device responsive to input of the fourth search query.

2. The method of claim 1, further comprising:

sending the second autocomplete result to the first search engine for input as a third search query; and

determining that a third autocomplete result from the first search engine responsive to input of the third search query is the same as the second autocomplete result.

3. The method of claim 1, further comprising:

determining that a first ranking value for the first autocomplete result satisfies an autocomplete ranking threshold; and

determining that a second ranking value for the second autocomplete result satisfies the autocomplete ranking threshold.

4. The method of claim 1, further comprising:

determining a first vector representation of the first autocomplete result, wherein the first vector representation comprises numerical values instead of letters;

determining a second vector representation of the second autocomplete result, wherein the second vector representation comprises numerical values instead of letters;

calculating a difference between individual values of the first vector representation and the second vector representation; and

determining that the difference satisfies a phrase difference threshold.

5. A method comprising:

identifying, by one or more computer processors, a set of seed words for input at a first search engine, the set of seed words comprising a first word;

determining a set of autocomplete results associated with the first word by recursively submitting autocomplete results to the first search engine until an autocomplete result matches the set of seed words, or until a predetermined number of results have been submitted to the first search engine, wherein the set of autocomplete results comprises a first autocomplete result and a second autocomplete result;

determining a second word and a first noun in the second autocomplete result;

determining a set of user interaction histories with a service provider that comprise queries input at the service provider or at a second search engine, the set of user interaction histories comprising a first user interaction history associated with a user account;

determining that the first user interaction history comprises a query with the second word and the first noun;

associating a user action with the query in a database;

determining that the query is input at the service provider or the second search engine at a user device; and

automatically causing the user action to be implemented at the user device responsive to input of the query.

6. The method of claim 5, wherein the query is a first query, the method further comprising:

determining a second user interaction history that comprises a second query with the first word and a second noun;

determining a second user action performed after the second query in the second user interaction history; and

associating the second user action with the second query in the database.

7. The method of claim 5, further comprising:

determining a total set of user interaction histories comprising the query; and

determining that a threshold number of the total set of user interaction histories comprise an indication that the user action was performed after input of the query.

8. The method of claim 5, further comprising:

determining that the first autocomplete result satisfies an autocomplete ranking threshold; and

determining that the second autocomplete result satisfies the autocomplete ranking threshold.

9. The method of claim 5, further comprising:

determining that the first autocomplete result comprises the first word;

determining that a third autocomplete result associated with the first word does not include the first word; and

discarding the third autocomplete result.

10. The method of claim 5, further comprising:

determining a prompt phrase comprising the first word;

wherein determining the first autocomplete result associated with the first word at the first search engine comprises determining the first autocomplete result associated with the prompt phrase at the first search engine.

11. The method of claim 5, further comprising:

determining that a third autocomplete result associated with the first word is similar to the first autocomplete result; and

discarding the third autocomplete result.

12. The method of claim 5, wherein determining the set of seed words for input at the first search engine comprises:

determining a first set of verbs;

determining a set of synonyms of the first set of verbs;

determining a second set of verbs in the set of synonyms; and

determining the set of seed verbs comprising the first set of verbs and the second set of verbs.

13. The method of claim 5, further comprising:

determining that a third autocomplete result associated with the second autocomplete result at the first search engine is the same as the second autocomplete result.

14. The method of claim 5, further comprising:

encoding the first autocomplete result as a first vector;

encoding the second autocomplete result as a second vector; and

determining a difference between the first vector and the second vector.

15. The method of claim 5, further comprising:

determining that the first autocomplete result is an actionable query.

16. A device comprising:

at least one memory that stores computer-executable instructions; and

at least one processor configured to access the at least one memory and execute the computer-executable instructions to;

determine a set of seed words for input at a first search engine, the set of seed words comprising a first word;

determine a set of autocomplete results associated with the first word by recursively submitting autocomplete results to the first search engine until an autocomplete result matches the set of seed words, or until a predetermined number of results have been submitted to the first search engine, wherein the set of autocomplete results comprises a first autocomplete result and a second autocomplete result;

determine a second word and a first noun in the second autocomplete result;

determine a set of user interaction histories that comprises queries input at a second search engine, the set of user interaction histories comprising a first user interaction history associated with a user account;

determine that the first user interaction history comprises a query with the second word and the first noun;

associate a user action with the query in a database;

determine that the query is input at the service provider or the second search engine at a user device; and

automatically cause the user action to be implemented at the user device responsive to input of the query.

17. The device of claim 16, wherein the at least one processor is further configured to access the at least one memory and execute the computer-executable instructions to:

determine a total set of user interaction histories comprising the query; and

determine that a threshold number of the total set of user interaction histories comprise an indication that the user action was performed after input of the query.

18. The device of claim 16, wherein the at least one processor is further configured to access the at least one memory and execute the computer-executable instructions to:

determine that the first autocomplete result comprises the first word;

determine that a third autocomplete result associated with the first word does not include the first word; and

discard the third autocomplete result.

19. A method comprising:

determining, by one or more computer processors coupled to at least one memory, a prompt phrase using a first seed verb of a first set of verbs that represents digital user actions;

sending the prompt phrase to a first search engine for input as a first search query;

determining a set of autocomplete results associated with the first word by recursively submitting autocomplete results to the first search engine until an autocomplete result matches an autocomplete result supplied on a current iteration, or until a predetermined number of results have been submitted to the first search engine, wherein the set of autocomplete results comprises a first autocomplete result and a second autocomplete result;

determining the first seed verb from the second autocomplete result;

determining that a search history associated with a user account for a second search engine comprises a third search query that has the first seed verb;

determining first user activity data associated with the user account for a first time interval after the third search query;

determining that the first user activity data comprises a user action of changing a device setting; and

causing a device setting control user interface to be presented at the user device responsive to input of the fourth search query.