System and method providing expert audience targeting

• US 10,915,973 B2
• Filed: 03/04/2015
• Issued: 02/09/2021
• Est. Priority Date: 03/05/2014
• Status: Active Grant

• Current Assignee: Transform Sr Brands, LLC (ESL Investments, Inc.)
• Original Assignee: Transform Sr Brands, LLC (ESL Investments, Inc.)
• Inventors: Yang, Jen Hao, Sponseller, Ryan C., Sridharan, Bharath, Chung, Eui, Lampert, Edward S.
• Primary Examiner(s): Thein, Marissa
• Assistant Examiner(s): Ramphal, Latasha D
• Application Number:
  US14/638,256
• Publication Number:
  US 20150254785A1
• Time in Patent Office: 2,169 Days
• Field of Search: 705 26
• CPC Class Codes: G06N 7/01 Probabilistic graphical models, e.g. probabilistic networks
• CPC Class Codes: G06Q 10/063112 Skill-based matching of a person or a group to a task
• CPC Class Codes: G06Q 30/0613 Third-party assisted
• CPC Class Codes: G06Q 50/01 Social networking

Abstract

A system and method for directing queries to the most suitable potential responders of an audience selected from a crowd-sourced population from which to request information, based on information such as query content, query context, timing, location, preferred supporting resource(s), and source of the query.

Claims

1. A method of operating a system for directing user queries to the most suitable potential responders of an audience selected from a crowdsource population from which to request information, the method comprising:

in a computer system having text classification circuitry, resource refinement circuitry, user interface circuitry, network interface circuitry, and memory configured to associate, in storage, a question received from a user via a user device and one or both of;

corresponding contextual and metadata information, wherein the metadata information can be added, via one or more tags, by the user, by a member of the crowdsource population, and by an automated tagging service, wherein the text classification circuitry, the resource refinement circuitry, the user interface circuitry, the network interface circuitry, and the memory are operable to support the system for directing user queries to the most suitable potential responders of the audience selected from the crowdsource population from which to request information;

labeling documents stored in a repository that is operatively coupled to the computer system;

training the text classification circuitry to identify particular expert groups based on labels corresponding to the labeled documents;

designating, via one or more tags, particular members of the crowdsource population as part of an expert group that is associated with one or more particular topics;

receiving location information from the user device based on a GPS signal received via GPS circuitry of the user device;

analyzing the user question via the trained text classification circuitry of the computer system, wherein the trained text classification circuitry is configured to perform a natural language analysis and analyze the user question based, in part, on the labels corresponding to the labeled documents stored in the repository;

assigning identifiers with probabilities, via the trained text classification circuitry, to the user question, wherein the trained text classification circuitry is configured to provide a corresponding probability of correctness of assignment of each of the identifiers to the user question;

storing the user question in the memory in association with corresponding contextual and metadata information;

identifying, automatically by the trained text classification circuitry, based on the question and one or both of the corresponding contextual and metadata information, one or more areas of subject matter to which the user question may relate;

generating, automatically by the trained text classification circuitry for each one of the one or more areas of subject matter, a probability that the user question relates to the one of the one or more areas of subject matter;

distributing, automatically by the resource refinement circuitry via the network interface circuitry, the user question to one or more selected members of the crowdsource population, wherein the one or more selected members of the crowdsource population are part of the expert group based on the one or more tags associated with the user question, wherein the distributing is based on the one or both of the corresponding contextual and metadata information, and based on the generated probabilities, and wherein the distributing is configurable to be based on the received location information from the user device based on the GPS signal received via the GPS circuitry of the user device;

receiving, via the network interface circuitry, one or more responses corresponding to the user question;

generating, by the user interface circuitry, a graphical user interface via which the one or more corresponding responses are presented to the user via the user device; and

adjusting, by one or both of the trained text classification circuitry and the resource refinement circuitry, one or both of the identifying and the distributing, according to feedback received from the user on the one or more corresponding responses to the user question.

2. The method according to claim 1, wherein, when the metadata information is added by the automated tagging service, the automated tagging service is configured to detect an image that is part of the received question and to add, via one or more tags, metadata information identifying content in the image.

3. The method according to claim 1, wherein the one or both of corresponding contextual and metadata information comprises one or both of personal information of the user stored in the memory and a geographic location of the user.

4. The method according to claim 1, wherein the distributing comprises:

selecting, by the resource refinement circuitry based on the generated probabilities, one or more members of the crowdsource population to prepare one or more corresponding responses to the user question.

5. The method according to claim 4, wherein the selecting comprises adjusting, by the resource refinement circuitry, how many members of the crowdsource population are selected during the selecting.

6. The method according to claim 4, wherein the adjusting comprises:

adjusting, by the trained text classification circuitry, the identifying, wherein;

the adjusting of the identifying is based on feedback received on the one or more corresponding responses to the user question; and

the feedback is from members of the crowdsource population not selected during the selecting.

7. The method according to claim 1, wherein the adjusting comprises:

adjusting, by the resource refinement circuitry, the distributing based on real-time information indicative of the availability of the individual members of the crowdsource population.

8. A system for directing queries to the most suitable potential responders of an audience selected from a crowdsource population from which to request information, the system comprising:

a computer system having text-classification circuitry, resource refinement circuitry, user interface circuitry, network interface circuitry, and memory configured to associate, in storage, a question received from a user via a user device and one or both of corresponding contextual and metadata information, wherein the text classification circuitry, the resource refinement circuitry, the user interface circuitry, the network interface circuitry, and the memory are operable to support the system for directing user queries to the most suitable potential responders of the audience selected from the crowdsource population from which to request information, wherein the metadata information can be added, via one or more tags, by the user, by a member of the crowdsource population, and by an automated tagging service, wherein the computer system designates, via one or more tags, particular members of the crowdsource population as part of an expert group that is associated with one or more particular topics, and wherein the computer system receives location information from the user device based on a GPS signal received via GPS circuitry of the user device,

wherein;

the memory is configured to store the user question in the memory in association with corresponding contextual and metadata information;

the text classification circuitry is configured to;

train the text classification circuitry based on documents that have been labeled and stored in a repository that is operatively coupled to the computer system, wherein the text classification circuit is trained to identify particular expert groups based on labels corresponding to the labeled documents;

analyze, by the trained text classification circuitry, the user question using a natural language analysis and the labels corresponding to the labeled documents stored in the repository;

assign identifiers with probabilities, by the trained text classification circuitry, to the user question and provide a corresponding probability of correctness of assignment of each of the identifiers to the user question;

identify, automatically by the trained text classification circuitry, based on the question and one or both of the corresponding contextual and metadata information, one or more areas of subject matter to which the user question may relate;

generate, automatically by the trained text classification circuitry, for each one of the one or more areas of subject matter, a probability that the user question relates to the one of the one or more areas of subject matter;

adjust, by the trained text classification circuitry, the identification according to feedback received from the user on the one or more corresponding responses to the user question;

the resource refinement circuitry is configured to;

distribute, automatically via the network interface circuitry, the user question to one or more selected members of the crowdsource population, wherein the one or more selected members of the crowdsource population are part of the expert group based on the one or more tags associated with the user question, wherein the distribution is based on the one or both of the corresponding contextual and metadata information, and based on the generated probabilities, and wherein the distribution is configurable to be based on the received location information from the user device based on the GPS signal received via GPS circuitry of the user device; and

adjust the distribution according to feedback received from the user on the one or more corresponding responses to the user question;

the network interface circuitry is configured to receive one or more responses corresponding to the user question; and

the user interface circuitry is configured to generate a graphical user interface via which the one or more corresponding responses are presented to the user via the user device.

9. The system according to claim 8, wherein, wherein, when the metadata information is added by the automated tagging service, the automated tagging service is configured to detect an image that is part of the received question and to add, via one or more tags, metadata information identifying content in the image.

10. The system according to claim 8, wherein the one or both of corresponding contextual and metadata information comprises one or both of personal information of the user stored in the memory and a geographic location of the user.

11. The system according to claim 8, wherein the distribution comprises:

selection, by the resource refinement circuitry based on the generated probabilities, of one or more members of the crowdsource population to prepare one or more corresponding responses to the user question.

12. The system according to claim 11, wherein the selecting comprises adjustment, by the resource refinement circuitry, how many members of the crowdsource population to select as part of the selection.

13. The system according to claim 11, wherein:

the adjustment comprises adjustment, by the trained text classification circuitry, of the identification;

the adjustment of the identification is based on feedback received on the one or more corresponding responses to the user question; and

the feedback is from members of the crowdsource population not selected as part of the selection.

14. The system according to claim 8, wherein:

the adjustment comprises adjustment, by the resource refinement circuitry, of the distribution; and

the adjustment of the distribution is based on real-time information indicative of the availability of the individual members of the crowdsource population.

15. A non-transitory computer-readable medium comprising executable instructions for causing a computer system having text classification circuitry, resource refinement circuitry, user interface circuitry, network interface circuitry, and memory configured to associate, in storage, a question received from a user via a user device and one or both of corresponding contextual and metadata information, to perform the steps of a method of operating a system for directing queries to the most suitable potential responders of an audience selected from a crowdsource population from which to request information, the steps comprising:

labeling documents stored in a repository that is operatively coupled to the computer system;

training the text classification circuitry to identify particular expert groups based on labels corresponding to the labeled documents;

designating, via one or more tags, particular members of the crowdsource population as part of an expert group that is associated with one or more particular topics;

receiving location information from the user device based on a GPS signal received via GPS circuitry of the user device;

analyzing the user question via the trained text classification circuitry of the computer system, wherein the trained text classification circuitry is configured to perform a natural language analysis and analyze the user question based, in part, on the labels corresponding to the labeled documents stored in the repository;

assigning identifiers with probabilities to the user question, wherein the trained text classification circuitry is configured to provide a corresponding probability of correctness of assignment of each of the identifiers to the user question;

storing the user question in the memory in association with corresponding contextual and metadata information, wherein the metadata information can be added, via one or more tags, by the user, by a member of the crowdsource population, and by an automated tagging service;

identifying, automatically by the trained text classification circuitry, based on the question and one or both of the corresponding contextual and metadata information, one or more areas of subject matter to which the user question may relate;

generating, automatically by the trained text classification circuitry for each one of the one or more areas of subject matter, a probability that the user question relates to the one of the one or more areas of subject matter;

distributing, automatically by the resource refinement circuitry via the network interface circuitry, the user question to one or more selected members of the crowdsource population, wherein the one or more selected members of the crowdsource population are part of the expert group based on the one or more tags associated with the user question, wherein the distributing is based on the one or both of the corresponding contextual and metadata information, and based on the generated probabilities, and wherein the distributing is configurable to be based on the received location information from the user device based on the GPS signal received via GPS circuitry of the user device;

receiving, via the network interface circuitry, one or more responses corresponding to the user question;

generating, by the user interface circuitry, a graphical user interface via which the one or more corresponding responses are presented to the user via the user device; and

adjusting, by one or both of the trained text classification circuitry and the resource refinement circuitry, one or both of the identifying and the distributing, according to feedback received from the user on the one or more corresponding responses to the user question.

16. The non-transitory computer-readable medium according to claim 15, wherein the one or both of corresponding contextual and metadata information comprises one or both of a geographic location of the user and personal information of the user stored in the memory.

17. The non-transitory computer-readable medium according to claim 15, wherein the steps comprise:

analyzing one or both of an Internet-based location and a GPS-based location information related to the user who submitted the user question, wherein one or both of the Internet-based location and the GPS-based location information are provided by circuitry of the user device;

determining, based on one or both of the Internet-based location and the GPS-based geographic information from the user device, an answer quality of one or more responses from the one or more selected members of the crowdsource population; and

adjusting one or both of the identifying and the distribution, according to the answer quality and the feedback received from the user on the one or more corresponding responses to the user question.

18. The non-transitory computer-readable medium according to claim 15, wherein the distributing comprises:

selecting, by the resource refinement circuitry based on the generated probabilities, one or more members of the crowdsource population to prepare one or more corresponding responses to the user question.

19. The non-transitory computer-readable medium according to claim 18, wherein the selecting comprises adjusting, by the resource refinement circuitry, how many members of the crowdsource population are selected during the selecting.

20. The non-transitory computer-readable medium according to claim 18, wherein the adjusting comprises:

adjusting, by the trained text classification circuitry, the identifying, wherein;

the adjusting of the identifying is based on feedback received on the one or more corresponding responses to the user question; and

the feedback is from members of the crowdsource population not selected during the selecting.

21. The non-transitory computer-readable medium according to claim 15, wherein the adjusting comprises:

adjusting, by the resource refinement circuitry, the distributing—

based on real-time information indicative of the availability of the individual members of the crowdsource population.