Expert system for designing experiments

• US 8,594,990 B2
• Filed: 12/29/2006
• Issued: 11/26/2013
• Est. Priority Date: 12/29/2005
• Status: Expired due to Fees

• Current Assignee: 3M Innovative Properties Company (3M Company)
• Original Assignee: 3M Innovative Properties Company (3M Company)
• Inventors: Brooks, Brian E.
• Primary Examiner(s): Thangavelu, Kandasamy
• Assistant Examiner(s): PIERRE LOUIS, ANDRE
• Application Number:
  US12/159,107
• Publication Number:
  US 20090281896A1
• Time in Patent Office: 2,524 Days
• Field of Search: 703/2, 703/20, 703/22, 705/14.4, 705/14.23, 705/14.38, 705/14.41, 705/14.65, 705/27.2
• US Class Current: 703/22
• CPC Class Codes: G06F 3/04842 Selection of displayed objects or displayed text elements G06F3/0482 takes precedence
• CPC Class Codes: G06F 40/186 Templates
• CPC Class Codes: G06N 5/04 Inference or reasoning models
• CPC Class Codes: G06Q 10/00 Administration; Management
• CPC Class Codes: G06Q 30/0245 Surveys
• CPC Class Codes: G06Q 30/0246 Traffic

Abstract

Computer implemented methods and systems for designing a true experiment, and optionally conducting the experiment and analyzing the experimental data are described. Information required to design the true experiment is identified. Questions are presented to a user eliciting responses that provide the required information. The required information is extracted from responses received from the user. The true experiment is designed based on the required information.

Claims

1. A computer-implemented method for designing a true experiment for evaluating effectiveness of a piece of content on influencing a particular behavior of viewers in a digital signage system having a plurality of displays, the method comprising:

identifying information required to design the true experiment to evaluate the piece of content, the identified information comprising information on a schedule confounding variable and a dependent variable that is associated with the behavior of the viewers of the piece of content, wherein the behavior of at least some of the viewers of the piece of content is exhibited independently from an interaction with a display in the digital signage system;

presenting questions to a user, the questions eliciting responses that provide the required information, wherein at least one of the questions is related to the timing of content that could produce carry over effects;

receiving the responses to the questions from the user;

extracting the required information from the user responses; and

designing, by a processor, the true experiment based on the required information extracted from the user responses,

wherein the designed experiment controls presentation of the piece of content on the plurality of displays in the digital signage system,

wherein the designed experiment controls for carry over effects by balancing and counterbalancing an order with which pieces of content are shown at locations across the digital signage system.

2. The method of claim 1, wherein designing the true experiment comprises determining a control group and a treatment group for the true experiment based on information extracted from the user responses, wherein the control group is exposed to the piece of content.

3. The method of claim 1, wherein designing the true experiment comprises determining a schedule for the piece of content based on the information extracted from the user responses.

4. The method of claim 3, wherein determining the schedule comprises determining one or more playlists for the piece of content based on the information extracted from the user responses.

5. The method of claim 3, wherein determining the schedule comprises determining one or more venues for the piece of content based on the information extracted from the user responses.

6. A system for designing a true experiment for evaluating effectiveness of a piece of content on influencing a particular behavior of viewers in a digital signage system having a plurality of displays, the system comprising:

a user interface configured to present questions to a user and receive user responses to the questions, wherein at least one of the questions is related to the timing of content that could produce carry over effects; and

an experimental design processor coupled to the user interface and configured to identify information required to perform a true experiment, generate questions eliciting responses from the user that provide the required information, extract the required information from the user responses, and design a true experiment using the required information,

wherein the identified information comprises information associated with a dependent variable that is associated with the behavior of the viewers of the piece of content, wherein the behavior of at least some of the viewers of the piece of content is exhibited independently from an interaction with a display in the digital signage system, and

wherein the designed experiment controls presentation of the piece of content on the plurality of displays in the digital signage system,

wherein the designed experiment controls for carry over effects by balancing and counterbalancing an order with which pieces of content are shown at locations across the digital signage system.

7. The system of claim 6, wherein the required information comprises information associated with one or more levels of an independent variable of the true experiment.

8. The system of claim 6, wherein the required information comprises information associated with one or both of confound variables and nuisance variables of the true experiment.

9. The system of claim 6, wherein the required information comprises information associated with a hypothesis of the true experiment.

10. The system of claim 6, wherein the true experiment evaluates digital signage content.

11. The system of claim 6, further comprising an experimental control processor configured to control performance of the true experiment.

12. The system of claim 6, further comprising an analysis unit configured to analyze experimental data produced by the true experiment.

13. The system of claim 12, wherein the analysis unit is configured to use information acquired by the design processor to analyze the experimental data.

14. A non-transitory computer-readable medium having instructions stored thereon and executable by a computing device to perform steps comprising:

identifying information required to design a true experiment for evaluating effectiveness of a piece of content on influencing a particular behavior of viewers in a digital signage system having a plurality of displays;

presenting questions to a user, the questions eliciting responses that provide the required information, wherein at least one of the questions is related to the timing of content that could produce carry over effects;

receiving the responses to the questions from the user;

extracting the required information from the user responses; and

designing the true experiment based on the required information extracted from the user responses,

wherein the identified information comprises information on a dependent variable that is associated with the behavior of the viewers of the piece of content, wherein the behavior of at least some of the viewers of the piece of content is exhibited independently from an interaction with a display in the digital signage system, and

wherein the designed experiment controls presentation of the piece of content on the plurality of displays in the digital signage system,

wherein the designed experiment controls for carry over effects by balancing and counterbalancing an order with which pieces of content are shown at locations across the digital signage system.

15. A system for designing a true experiment for evaluating effectiveness of a piece of content on influencing a particular behavior of viewers in a digital signage system having a plurality of displays, the system comprising:

means for identifying information required to design a true experiment;

means for presenting questions to a user, the questions eliciting responses that provide the required information, wherein at least one of the questions is related to the timing of content that could produce carry over effects;

means for receiving the responses to the questions from the user;

means for extracting the required information from the user responses; and

means for designing the true experiment based on the required information extracted from the user responses,

wherein the identified information comprises information on a dependent variable that is associated with the behavior of the viewers of the piece of content, wherein the behavior of at least some of the viewers of the piece of content is exhibited independently from an interaction with a display in the digital signage system, and

wherein the designed experiment controls presentation of the piece of content on the plurality of displays in the digital signage system,

wherein the designed experiment controls for carry over effects by balancing and counterbalancing an order with which pieces of content are shown at locations across the digital signage system.

16. The system of claim 15, further comprising means for conducting the true experiment.

17. The system of claim 15, further comprising means for analyzing experimental data produced by the true experiment.