Ranking suggestions based on user attributes

Patent metadata

Field	Value
Publication number	US-9805142-B2
Application number	US-201414250504-A
Country	US
Kind code	B2
Filing date	Apr 11, 2014
Priority date	Apr 11, 2014
Publication date	Oct 31, 2017
Grant date	Oct 31, 2017

Abstract

Methods and apparatus for ranking suggestions for input of a user based on one or more attributes of the user that are associated with one or more of the suggestions. User attributes of the user's user profile may be matched to user attributes of a suggestion and the suggestion ranked based at least in part on bias measures associated with the matched user attributes of the suggestion. Some implementations are directed to determining, for each of a plurality of suggestions, multiple user attributes and associated bias measures to associate with the suggestion.

First claim

What is claimed is:

1. A computer-implemented method, comprising:
identifying suggestions for an input of a user, each of the suggestions including one or more characters not present in the input;
identifying user attributes associated with the user;
identifying, for a given suggestion of the suggestions, a set of the user attributes that are each associated with the given suggestion;
identifying, for each of the user attributes of the set, a given suggestion bias measure for the given suggestion, wherein the given suggestion bias measure for a given user attribute of the set is based on a measure of submissions associated with the given suggestion by a set of users having indicated interactions associated with the given user attribute;
determining an overall bias measure for the given suggestion based on the given suggestion bias measures for the user attributes of the set;
determining a ranking of the given suggestion for the input based on the overall bias measure;
identifying a submission of the given suggestion by the user;
identifying additional suggestions for a second input of the user, wherein the second input is subsequent to the submission of the given suggestion;
identifying, for a second given suggestion of the additional suggestions, a second set of the user attributes associated with the user that are also each associated with the second given suggestion, wherein the user attributes of the second set include overlapping attributes that are included in both the set, associated with the given suggestion, and the second set, associated with the second given suggestion;
identifying, for each of the overlapping attributes, a second bias measure for the second given suggestion, wherein the second bias measures for the overlapping attributes vary from the given suggestion bias measures for the overlapping attributes;
determining a second overall bias measure for the second given suggestion, wherein the second overall bias measure is based on the second bias measures for the overlapping attributes and the given suggestion bias measures for the overlapping attributes, the given suggestion bias measures for the overlapping attributes being bias measures for the given suggestion and based on the measures of submissions associated with the given suggestion, and wherein the given suggestion bias measures are used based on the second input of the user being subsequent to the previously submitted given suggestion;
determining a ranking of the second given suggestion for the second input based on the second overall bias measure; and
providing the second given suggestion for presentation as a suggestion for the second input, wherein providing the second given suggestion is based on the ranking.

2. The computer-implemented method of claim 1 , wherein the bias measure for the given user attribute of the set is based on a measure of the indicated interactions associated with the given user attribute by the set of users.

3. The computer-implemented method of claim 2 , wherein the given user attribute of the set is associated with one or more documents or locations and wherein the indicated interactions associated with the given user attribute includes interactions by the set of users with the one or more documents or locations.

4. The computer-implemented method of claim 3 , wherein the measure of the indicated interactions is based on one or more popularity measures of the one or more documents or locations associated with the given user attribute.

5. The computer-implemented method of claim 3 , wherein the measure of the indicated interactions by the set of users is based on comparison of the indicated interactions to additional indicated interactions, the additional indicated interactions including interactions with additional documents or locations that are not included in the one or more documents or locations.

6. The computer-implemented method of claim 1 , wherein the given user attribute is included in the overlapping attributes, and wherein the second bias measure for the given user attribute is based on a measure of submissions associated with the second given suggestion by the set of users having the indicated interactions associated with the given user attribute.

7. The computer-implemented method of claim 1 , wherein the second bias measures for the second given suggestion and the bias measures for the overlapping attributes are weighted evenly in determining the second overall bias measure for the second given suggestion.

8. The computer-implemented method of claim 1 , wherein the second bias measures for the second given suggestion are weighted more heavily than the bias measures for the overlapping attributes in determining the second overall bias measure for the second given suggestion.

9. The computer-implemented method of claim 1 , further comprising:
identifying, for a second suggestion of the suggestions, an additional set of one or more of the user attributes that are each associated with the second suggestion;
identifying, for each of the user attributes of the additional set, a second suggestion bias measure for the second suggestion;
determining a second suggestion overall bias measure for the second suggestion based on the second suggestion bias measures for the user attributes of the additional set; and
determining a ranking of the second suggestion based on the second suggestion overall bias measure.

10. The computer-implemented method of claim 9 , wherein one or more of the user attributes of the set are not present in the additional set.

11. The computer-implemented method of claim 9 , wherein the set and the additional set include one or more same of the user attributes.

12. The computer-implemented method of claim 1 , wherein the input is a partial query.

13. A computer-implemented method, comprising:
identifying a submitted query submitted by a user;
identifying user attributes associated with the user;
identifying, for the submitted query, a set of the user attributes that are each associated with the submitted query;
identifying, for each of the user attributes of the set, a submitted bias measure for the submitted query, wherein the submitted bias measure for a given user attribute of the set is based on a measure of submissions associated with the submitted query by a set of users having the given user attribute;
identifying query suggestions for a second query of the user, wherein the second query is subsequent to the submitted query and wherein each of the query suggestions includes one or more characters not present in the submitted query;
identifying, for a given query suggestion of the query suggestions, a second set of the user attributes associated with the user that are also each associated with the given query suggestion, wherein the user attributes of the second set include overlapping attributes, the overlapping attributes being included in both the set, associated with the submitted query, and the second set, associated with the given query suggestion;
identifying, for each of the overlapping attributes, a given suggestion bias measure for the given query suggestion, wherein the given suggestion bias measures for the overlapping attributes vary from the submitted bias measures for the overlapping attributes;
determining an overall bias measure for the given query suggestion, wherein determining the overall bias measure is based on the given suggestion bias measures and the submitted bias measures for the overlapping attributes, the submitted bias measures for the overlapping attributes being bias measures for the submitted query and based on submissions associated with the submitted query, and wh

Assignees

• Google Inc

Inventors

• Sanghai Sumit Kumar
• Sivakumar Dandapani
• Tiwari Pranav Kumar
• Keshavan Raghunandan Hulikal

Classifications

• G06F17/30867

  Physics · mapped topic

• G06F17/3064

  Physics · mapped topic

• G06F17/30973 Primary

  Physics · mapped topic

• G06F16/3322 Primary

  using system suggestions (G06F16/3325 takes precedence) · CPC title

• G06F16/90328 Primary

  using search space presentation or visualization, e.g. category or range presentation and selection · CPC title