Method and apparatus for automatic selection and presentation of sales promotion programs

• US 5,504,675 A
• Filed: 12/22/1994
• Issued: 04/02/1996
• Est. Priority Date: 12/22/1994
• Status: Expired due to Fees

• Current Assignee: International Business Machines Corporation
• Original Assignee: International Business Machines Corporation
• Inventors: Lund, Stephen H., Kelsey, Todd M., Cragun, Brian J.
• Primary Examiner(s): Hayes, Gail O.
• Assistant Examiner(s): Poinvil, Frantzy
• Application Number:
  US08/363,195
• Time in Patent Office: 467 Days
• Field of Search: 364/400, 364/401, 364/405, 364/479, 340/825.33, 340/825.35, 340/541, 235/379, 235/380, 395/2.68, 395/2.79, 395/2.83, 395/23, 395/925
• US Class Current: 705/14.4
• CPC Class Codes: G06Q 30/02 Marketing; Price estimation or determination; Fundraising
• CPC Class Codes: G06Q 30/0241 Advertisements

Abstract

A sales promotion program is dynamically selected from a plurality of programs for presentation in a program presentation unit by a neural network that makes its selection based on first detecting if a person is in the area immediately around the program presentation unit, then either selecting a general attract loop sales promotion program with the trained neural network using a set of predetermined system criteria if no person is detected in the immediate area or selecting a specific loop sales promotion program if at least one person is detected in the immediate area. The neural network is trained by selecting general attract loop programs that are run and then collecting data indicative of the number of persons responding to the general attract loop and also by selecting specific loop programs that are run if a person is in the immediate area and then collecting data indicative of the responses to the specific loop programs. The collected data thereby represents the success of the various sales programs in attracting and holding the attention of persons. The collected data is provided to the neural network in any one of a plurality of training schemes typical for neural networks, after which the trained neural network is provided with current, real-time selection data such that the trained network can select the most appropriate sales promotion program for running. The network can be retrained at regular intervals or in response to sales data or changes in the collected data.

Claims

1. A presentation unit comprising:

a display touch screen unit that includes a touch screen on which images can be displayed, receives display image information from a sales promotion program, and produces a touch signal in response to a touch at the touch screen;

at least one positive response sensor that produces a signal that indicates when a person is in an area physically near the presentation unit and responds to the displayed images in a positive manner;

a speaker that transduces sound front the sales promotion program; and

a central processing unit having one of said neural network selection program and a plurality of sales promotion programs, wherein the neural network selection program selects one of said sales promotion program for presentation in response to real-time data and signals from the positive response sensor, said presentation of the selected program being displayed by the display touch screen and playbacked by the speaker.

2. A presentation unit as defined in claim 1, wherein the positive response sensor includes a proximity sensor that produces a proximity signal that indicates when a person is in the immediate area of the touch screen and the positive manner of responding comprises a person touching the touch screen.

3. A presentation unit as defined in claim 2, wherein the positive response sensors further include a traffic sensor that produces a traffic index signal that indicates the number of persons in the immediate area of the presentation unit.

4. A presentation unit as defined in claim 2, wherein the central processing unit further contains:

a set of collected selection data for training the neural network selection program, the selection data including traffic volume data that indicates the number of persons in the immediate area of the touch screen at predetermined times in accordance with randomly selected sales promotion programs presented at the predetermined times.

5. A presentation unit as defined in claim 4, wherein the sales promotion programs include general attract loop sales promotion programs and specific loop sales promotion programs.

6. A presentation unit as defined in claim 5, wherein the set of collected selection data comprises a plurality of data records, each of which includes the current date, current time of day, an index of current traffic in the immediate area, and an index of general attract loop successes at predetermined times.

7. A presentation unit as defined in claim 6, wherein the neural network is trained using a neural network training epoch data subset of the collected selection data comprising a predetermined number of data records.

8. A presentation unit as defined in claim 7, wherein the central processing unit generates a neural network output set of program indexes in response to the collected input data and the general attract loop sales promotion program is selected by the central processing unit in response to the program index having the highest output value.

9. A presentation unit as defined in claim 8, wherein the central processing unit selects a specific loop sales promotion program only after the touch screen of the program presentation unit was touched by a person.

10. A presentation unit as defined in claim 9, wherein the specific loop sales promotion program is selected only if the display touch screen was touched within a predetermined time interval.

11. A presentation unit as defined in claim 6, wherein an attract loop success comprises the number of persons detected in the immediate area of the presentation unit and a specific loop success comprises the number of touch screen interactions in a predetermined time interval.

12. A kiosk unit that presents multimedia sales promotion programs having audio and video programming, the kiosk unit comprising:

a kiosk housing;

a display touch screen unit that is located in the kiosk housing and includes a touch screen on which images can be displayed so the displayed images are visible to passers-by, receives display image information from a sales promotion program, and produces a touch signal in response to a touch at the touch screen;

at least one proximity sensor that produces a proximity signal that indicates when a person is in an immediate area of the touch screen;

a traffic sensor that produces a traffic index signal that indicates the number of persons in the immediate area of the kiosk housing;

a time data subsystem that generates time data comprising date and time-of-day data;

a speaker that transduces sound from the sales promotion programs during playback so the transduced sound can be heard by passers-by; and

a central processing unit having a neural network selection program and a plurality of sales promotion programs, wherein the neural network selection program assumes an untrained state in which selection data is collected and assumes a trained state in which the neural network selection program selects a sales promotion program for presentation whenever it is not currently presenting a sales promotion program in accordance with collected selection data said central processing means including means for outputting a presentation of selected sales promotion programs to the touch screen and a playback back to the speaker in response to real-time input data including the time data and collected data signals from the touch screen unit, proximity sensor, and traffic sensor according to said time data subsystem.

13. A kiosk unit as defined in claim 12, wherein the set of collected selection data comprises a plurality of data records, each of which includes the current date, current time of day, an index of current traffic in the immediate area, and an index of general attract loop successes at predetermined times and the collected selection data further includes traffic volume data that indicates the number of persons in the immediate area of the touch screen according to the time data.

14. A kiosk unit as defined in claim 13, wherein the sales promotion programs include general attract loop sales promotion programs and specific loop sales promotion programs.

15. A kiosk unit as defined in claim 14, wherein the central processing unit generates a neural network output set of program indexes in response to the collected input data and the general attract loop sales promotion program is selected by the central processing unit in response to the greatest program index.

16. A kiosk unit as defined in claim 15, wherein the central processing unit selects a specific loop sales promotion program only after the touch screen of the kiosk unit was touched by a person.

17. A kiosk unit as defined in claim 16, wherein the specific loop sales promotion program is selected only if the display touch screen was touched within a predetermined time interval.

18. A kiosk unit as defined in claim 13, wherein an attract loop success comprises the number of persons detected in the immediate area of the kiosk unit and a specific loop success comprises the number of touch screen interactions that occur in a predetermined time interval.

19. A method of dynamically selecting one of a plurality of sales promotion programs for running in a program presentation unit, the method comprising the steps of:

training a neural network that generates an output data set indicating one of the plurality of sales promotion programs to be run in the program presentation unit along with a set of selection data;

providing the trained neural network with current selection data;

detecting if a person is in the area immediately around the program presentation unit;

selecting one of a plurality of general attract loop sales promotion programs with the trained neural network using a set of predetermined system criteria, for running in the program presentation unit, if no person is detected in the immediate area;

selecting a specific loop sales promotion program with the trained neural network in response to real-time data, for running in the program presentation unit, if at least one person is detected in the immediate area; and

running the selected sales promotion program in the program presentation unit.

20. A method as defined in claim 19, wherein the step of training the neural network comprises the steps of:

detecting if a person is in an area immediately around the program presentation unit;

randomly selecting a general attract loop sales promotion program if no person is detected in the immediate area;

randomly selecting a sales loop sales promotion program if at least one person is detected in the immediate area;

running the randomly selected program in the program presentation unit;

collecting a set of selection data for training the neural network, the selection data including data relating to the traffic and environment in the immediate area;

repeating the steps of detecting, selecting, and running for a predetermined time interval; and

providing the neural network with the collected selection data.

21. A method as defined in claim 20, wherein the step of collecting includes collecting selection data comprising the current date, current time of day, an index of current traffic in the immediate area, and an index of attract loop successes comprising the number of persons detected in the immediate area of the presentation unit.

22. A method as defined in claim 20, wherein the step of providing comprises:

obtaining a first neural network training epoch data subset of the collected selection data;

performing a neural network training process with the data subset of the collected selection data;

obtaining a next neural network training epoch data subset of the collected selection data; and

repeating the steps of performing and obtaining until all neural network training epoch data subsets in the collected selection data have been processed.

23. A method as defined in claim 19, wherein the step of selecting a general attract loop sales promotion program comprises the steps of:

collecting a set of selection input data for the neural network;

providing the collected input data to the neural network and generating a neural network output set of program indexes; and

selecting the general attract loop sales promotion program corresponding to the greatest program index.

24. A method as defined in claim 19, wherein the step of selecting a sales loop sales promotion program further comprises the step of:

selecting a specific loop sales promotion program with the trained neural network, for running in the program presentation unit, if a touch screen of the program presentation unit was touched by a person.

25. A method as defined in claim 24, wherein the specific loop sales promotion program is selected only if the touch screen was touched within a predetermined time interval.