Identification of query completions which change users' original search intent

• US 9,355,191 B1
• Filed: 01/24/2013
• Issued: 05/31/2016
• Est. Priority Date: 01/24/2012
• Status: Active Grant

• Current Assignee: Google LLC (Alphabet Inc.)
• Original Assignee: Google Inc. (Alphabet Inc.)
• Inventors: Eiger, Sivan, Finkelstein, Lev
• Primary Examiner(s): Syed, Farhan
• Application Number:
  US13/748,929
• Time in Patent Office: 1,223 Days
• US Class Current: 1/1
• CPC Class Codes: G06F 16/90324 using system suggestions

Abstract

Technology described herein identifies and demotes query completions which divert users from their original search intent. The identified query completions may represent informational requests that are amusing diversions for some users, but which are not relevant to the actual information of interest for many users. For example, the identified query completions may represent information requests that some users may find humorous or entertaining, but which are not relevant to actual information of interest for many other users. The identification of these query completions may allow alternative query completions to be provided that can more likely assist users in finding the information they actually seek.

Claims

1. A computer-implemented method performed by one or more computers, the method comprising:

calculating, by one or more of the computers, a completion origination score for a query completion based on a frequency of client computing device submission of the query completion as a search query upon being provided as a suggested query completion and a total frequency of client computing device submission of the query completion as a search query;

identifying, by one or more of the computers based on the completion origination score, the query completion as potentially changing users'"'"' original search intent;

testing, by one or more of the computers, an impact of withholding the identified query completion on a frequency of client computing device submission of search queries matching the identified query completion, wherein the testing is in response to identifying the query completion as potentially changing the users'"'"' original search intent, and wherein the testing comprises preventing the identified query completion from being provided as a suggested completion in response to at least one corresponding partial query submitted by at least some client computing devices of users during the testing; and

classifying, by one or more of the computers, the identified query completion as search intent changing based on a decrease in the frequency of client computing device submission of search queries matching the identified query completion, the decrease being in a frequency of client computing device submission of search queries entered by users during the testing which match the identified query completion relative to a frequency of client computing device submission of search queries matching the identified query completion during a time interval outside of the testing;

in response to classifying the identified query completion as search intent changing, demoting, by one or more of the computers, a popularity score of the identified query completion, the popularity score being assigned to the identified query completion in a database and being used to select the identified query completion as a suggested completion for partial queries submitted by users.

2. The method of claim 1, wherein the identified query completion is prevented from being provided as a suggested completion in response to the at least one corresponding partial query submitted by a predetermined percentage of users during the testing.

3. The method of claim 1, wherein classifying the identified query completion as search intent changing is based on a frequency of client computing device submission of search queries matching the identified query completion entered by users when the identified query completion has been withheld during the testing.

4. The method of claim 1, wherein the search queries matching the identified query completion are identical to the identified query completion.

5. The method of claim 1, wherein classifying the identified query completion as search intent changing comprises:

estimating a confidence interval based on the decrease in the frequency of client computing device submission of search queries matching the identified query completion during the testing;

determining that the confidence interval satisfies a threshold; and

in response to the determination, classifying the identified query completion as search intent changing.

6. The method of claim 1, wherein the identified query completion is a popular query which been submitted more than a threshold number of times.

7. A non-transitory computer readable storage medium storing computer instructions executable by a processor to perform a method comprising:

calculating a completion origination score for a query completion based on a frequency of client computing device submission of the query completion as a search query upon being provided as a suggested query completion and a total frequency of client computing device submission of the query completion as a search query;

identifying, based on the completion origination score, the query completion as potentially changing users'"'"' original search intent;

testing an impact of withholding the identified query completion on a frequency of client computing device submission of search queries matching the identified query completion, wherein the testing is in response to identifying the query completion as potentially changing the users'"'"' original search intent, and wherein the testing comprises preventing the identified query completion from being provided as a suggested completion in response to at least one corresponding partial query submitted by at least some client computing devices of users during the testing; and

classifying the identified query completion as search intent changing based on a decrease in the frequency of client computing device submission of search queries matching the identified query completion, the decrease being in a frequency of client computing device submission of search queries entered by users during the testing which match the identified query completion relative to a frequency of client computing device submission of search queries matching the identified query completion during a time interval outside of the testing;

in response to classifying the identified query completion as search intent changing, demoting a popularity score of the identified query completion, the popularity score being assigned to the identified query completion in a database and being used to select the identified query completion as a suggested completion for partial queries submitted by users.

8. The non-transitory computer readable storage medium of claim 7, wherein the identified query completion is prevented from being provided as a suggested completion in response to the at least one corresponding partial query submitted by a predetermined percentage of users during the testing.

9. The non-transitory computer readable storage medium of claim 7, wherein classifying the identified query completion as search intent changing is based on a frequency of client computing device submission of search queries matching the identified query completion entered by users when the identified query completion has been withheld during the testing.

10. The non-transitory computer readable storage medium of claim 7, wherein classifying the identified query completion as search intent changing comprises:

estimating a confidence interval based on the decrease in the frequency of client computing device submission of search queries matching the identified query completion during the testing;

determining that the confidence interval satisfies a threshold; and

in response to the determination, classifying the identified query completion as search intent changing.

11. A system including memory and one or more processors operable to execute instructions, stored in the memory, comprising instructions to:

calculate a completion origination score for a query completion based on a frequency of client computing device submission of the query completion as a search query upon being provided as a suggested query completion and a total frequency of client computing device submission of the query completion as a search query;

identify, based on the completion origination score, the query completion as potentially changing users'"'"' original search intent;

test an impact of withholding the identified query completion on a frequency of client computing device submission of search queries matching the identified query completion, wherein the testing is in response to identifying the query completion as potentially changing the users'"'"' original search intent, and wherein the testing comprises preventing the identified query completion from being provided as a suggested completion in response to at least one corresponding partial query submitted by at least some client computing devices of users during the testing; and

classify the identified query completion as search intent changing based on a decrease in the frequency of client computing device submission of search queries matching the identified query completion, the decrease being in a frequency of client computing device submission of search queries entered by users during the testing which match the identified query completion relative to a frequency of client computing device submission of search queries matching the identified query completion during a time interval outside of the testing;

demote, in response to classifying the identified query completion as search intent changing, a popularity score of the identified query completion, the popularity score being assigned to the identified query completion in a database and being used to select the identified query completion as a suggested completion for partial queries submitted by users.

12. The system of claim 11, wherein the identified query completion is prevented from being provided as a suggested completion in response to the at least one corresponding partial query submitted by a predetermined percentage of users during the testing.

13. The system of claim 11, wherein the instructions to classify the identified query completion as search intent changing is based on a frequency of client computing device submission of search queries matching the identified query completion entered by users when the identified query completion has been withheld during the testing.

14. The system of claim 11, wherein the instructions to classify the identified query completion as search intent changing comprises instructions to:

estimate a confidence interval based on a decrease in the frequency of client computing device submission of search queries matching the identified query completion during the testing;

determine that the confidence interval satisfies a threshold; and

in response to the determination, classify the identified query completion as search intent changing.