Discrete choice method of reporting and predicting multiple transaction types

• US 8,184,547 B2
• Filed: 11/18/2004
• Issued: 05/22/2012
• Est. Priority Date: 11/18/2004
• Status: Active Grant

• Current Assignee: Alvaria Cayman
• Original Assignee: Aspect Software Incorporated
• Inventors: Sumner, Roger, Dezonno, Anthony
• Primary Examiner(s): Hailu, Kibrom T
• Application Number:
  US10/991,757
• Publication Number:
  US 20060104213A1
• Time in Patent Office: 2,742 Days
• Field of Search: 370241-253, 370/485, 370/335, 370/352, 370/264, 370/265, 370/341, 370/433, 370/480, 707/3, 709/223, 705/8, 705/10, 705/7
• US Class Current: 370/252
• CPC Class Codes: H04L 41/00 Arrangements for maintenance, administration or management of data switching networks, e.g. of packet switching networks
• CPC Class Codes: H04L 41/142 using statistical or mathematical methods
• CPC Class Codes: H04L 41/145 involving simulating, designing, planning or modelling of a network
• CPC Class Codes: H04L 41/147 for predicting network behaviour
• CPC Class Codes: H04L 41/34 Signalling channels for network management communication

Abstract

The method for and system or apparatus for forecasting future communication transaction traffic from a customer include the steps of or structure for: collecting communication channel data on at least first and second communication channels; performing a numeric transformation to the data; calculating a discrete choice probability for each communication channel; and forecasting future period usage values for a user on a communication channel.

Claims

1. A method performed by a processor for forecasting future communication transaction traffic from a customer comprising the steps of:

a) collecting communication channel data of communication channel selections by the customer for at least first and second communication channels, each communication channel using a different communication method, and communication each communication channel available to be chosen by the customer for communications;

b) performing a numeric transformation to the data;

c) calculating a discrete choice probability of communication channel selection by the customer for each communication channel available to be chosen by the customer using discrete choice modeling; and

d) forecasting future channel selection trends of communication channel selections to be made by the customer on at least one of the communication channels based on the discrete choice probabilities.

2. The method of claim 1 wherein the first communication channel is selected from one of a voice telephone number, a data telephone number, an email address, a sip address, an uniform resource locator (URL) address, an Uniform Resource Identifier (URI) address, an Internet Protocol (IP) address, an instant messaging identifier, and a cable box identifier.

3. The method of claim 1 wherein the second communication channel is selected from one of a voice telephone number, a data telephone number, an email address, a sip address, an uniform resource locator (URL) address, an Uniform Resource Identifier (URI) address, an Internet Protocol (IP) address, an instant messaging identifier, and a cable box identifier.

4. The method of claim 1 wherein the numeric transformation uses at least one of an exponential formula, a Hayes transformation, or a logit function.

5. The method of claim 1 wherein the calculation of a discrete choice probability results from an arithmetic operation of the numeric transforms.

6. The method of claim 1 wherein the calculation of a discrete choice probability is performed using multimedia logit estimation.

7. The method of claim 1 wherein the forecasting of future communication channel selection trends uses a least mean squares formula.

8. The method of claim 1 wherein the forecast is displayed on one of a printout from a printer or a visual display.

9. The method of claim 1, wherein the discrete choice probability of a communications channel is pA|[A,B], and

pA|[A,B]=UA/(UA+UB)

where UA is an exponential transform value of a first communication channel value;

where UB is an exponential transform value of a second communication channel value; and

where (UA+UB) is a sum of the exponential transforms of multiple communication channels.

10. The method of claim 1, wherein the discrete choice probability of a communications channel equals 1, and

pA+pB=1

wherein pA is the probability of use of a first communication channel value;

and

where pB is the probability of use of a second communication channel value.

11. A non-transitory computer-readable storage media embodying a program of instructions executable by a computer to perform the steps of:

a) collecting communication channel selection data for a user on at least two communication channels;

b) performing a numeric transformation to the data;

c) calculating a discrete choice probability for the user of user communication channel selection of each communication channel using discrete choice modeling; and

d) forecasting future communication channel selection trends for communication channel selections to be made by the user on the communication channels based upon the discrete choice probabilities.

12. The non-transitory computer-readable storage media of claim 11 wherein said media comprise at least one of a random access memory, a read only memory, a disk, an application specific integrated circuit, and a programmable read only memory.

13. The non-transitory computer-readable storage media tangibly embodying a program of instructions executable by a computer to perform the steps of:

collecting communication channel user choice data on at least two communication channels available to be chosen by users for a communication;

performing a numeric transformation to the data;

calculating a discrete choice probability for user communication channel choices by the users for each communication channel using discrete choice modeling; and

forecasting future period values for the communication channels based upon the discrete choice probabilities.

14. The non-transitory computer-readable storage media of claim 13 wherein said media comprise at least one of a random access memory, a read only memory, a disk, an application specific integrated circuit, and a programmable read only memory.

15. A processor for forecasting future communication transaction traffic from a customer comprising:

means for collecting communication channel selection data of communication channel selections made by the customer on at least two communication channels;

means for performing a numeric transformation to the data;

means for calculating a discrete choice probability of communication channel selection by the customer for each communication channel using discrete choice modeling; and

means for forecasting future communication channel selection values of communication channel selection made by the customer on at least one of the communication channels based on the discrete choice probabilities.

16. A processor for forecasting future communication transaction traffic from a customer comprising:

means for collecting communication channel selection data of communication channel selections by the customer on at least two communication channels;

means for performing a numeric transformation to the data;

means for calculating a discrete choice probability of communication channel selections for each communication channel; and

means for forecasting future period communication channel selection trends of the customer for the communication channels based on the discrete choice probabilities.

17. An apparatus programmed to forecast future communication transaction traffic comprising:

a communications interface to a transaction processing system including at least one processor, wherein said communication interface provides communication channel traffic level information about a plurality of available communications channels, each of the plurality of communications channels using different communications method, to a computing system which determines discrete choice probabilities of communication channel selection by a customer for each channel based upon the traffic information; and

a processing system, said processing system being disposed within the computing system for forecasting future trend values for channel selection by the customer based on the discrete choice probabilities for at least one of the plurality of communication channels.

18. An apparatus programmed to forecast future communication transaction traffic comprising:

a communications interface to a transaction processing system including at least one processor, wherein said communication interface provides communication traffic level information about a user of a plurality of available communications channels, each of the plurality of communication channels using a different communications method, to a computing system; and

a processing system, said processing system being disposed within the computing system for forecasting future trend values for communication channel selection by the user of at least one of the plurality of communication channels based upon discrete choice probabilities of communication channel selection by the user derived from the traffic level information.

19. A computer programmed to execute a process of forecasting communication transaction traffic, the process comprising:

collecting communication channel selection data of channel selections by a customer on at least two communication channels;

performing a numeric transformation to the data; and

calculating a discrete choice probability of communication channel selection by the customer for user communication channel selection of each communication channel; and

forecasting communication transaction traffic using the discrete choice probability of channel selection.

20. The computer of claim 19, wherein the process further comprises:

forecasting future communication channel selection values for the customer selection of a communication channel.

21. The computer of claim 19, wherein the process further comprises:

forecasting a probability of next communication channel selection by the customer of a communication channel based upon the discrete choice probabilities.

22. The computer to execute a process of forecasting communication transaction traffic of claim 19 wherein the collecting of communication channel data further comprises defining the communication channel as being a telephone number of an automatic contact distributor and the data as being call traffic level of the telephone number of the automatic contact distributor.

23. The computer of claim 19, wherein the discrete choice probability of a communications channel is pA|[A,B], and

pA|[A,B]=UA/(UA+UB)

where UA is exponential transform value of a first communication channel value;

where UB is the exponential transform value of a second communication channel value; and

where (UA+UB) is sum of the numeric transforms of multiple communication channels.

24. The computer of claim 19, wherein the probability choosing communications channel equals 1, and

pA+pB=1

wherein pA is the probability of use of a first communication channel value;

and

where pB is the probability of use of a second communication channel value.

25. A method performed by a processor for forecasting future communication transaction traffic from a customer including the steps of:

collecting over a selected time period N versions of communication transaction method selection data including at least two different communication transaction methods, the first method of communication transactions being identified as Type A mad the second method of communication transactions being identified as Type B;

tabulating the data collected;

performing a numeric transformation to the data;

calculating a discrete choice probability of communications method selection by the customer for each communication method using discrete choice modeling;

forecasting future trends of communication method selection made by the customer for each of the methods of communications transactions being used based on the discrete choice probabilities; and

displaying the forecast.

26. The method of claim 25 wherein the Type A transaction uses one of a voice telephone number, a data telephone number, an email address, a sip address, an uniform resource locator (URL) address, an Uniform Resource Identifier (URL) address, an Internet Protocol (IP) address, an instant messaging identifier, or a cable box identifier.

27. The method of claim 25 wherein the Type B transaction uses one of a voice telephone number, a data telephone number, an email address, a sip address, an uniform resource locator (URL) address, an Uniform Recourse Identifier (URL) address, an Internet Protocol (IP) address, an instant messaging identifier, or a cable box identifier.

28. The method of claim 25 wherein the discrete choice modeling uses multimedia logit estimation.

29. The method of claim 25 wherein the numeric transformation uses an exponential formula.

30. The method of claim 25 wherein the numeric transformation uses Hayes modeling.

31. A system for forecasting future communication transaction traffic from a customer comprising:

hardware and software for collecting over a selected time period N versions of communication transaction method selection data including at least two different communication transaction methods, the first method of communication transactions being identified as Type A and the second method of communication transactions being identified as Type B;

hardware and software for tabulating the data collected;

hardware and software for performing a numeric transformation to the data;

hardware and software for calculating a discrete choice probability for communication method selection by the customer for each communication method using discrete choice modeling;

hardware and software for forecasting future probability values of selection by customer for each of the methods of communication transactions being used based upon the discrete choice probabilities; and

display structure for displaying the forecast.

32. The system of claim 31 wherein the Type A transaction uses one of a voice telephone number, a data telephone number, an email address, a sip address, an uniform resource locator (URL) address, an Uniform Resource Identifier (URI) address, an Internet Protocol (IP) address, an instant messaging identifier, or a cable box identifier.

33. The system of claim 31 wherein the Type B transaction uses one of a voice telephone number, a data telephone number, an email address, a sip address, an uniform resource locator (URL) address, an Uniform Resource Identifier (URI) address, an Internet Protocol (IP) address, an instant messaging identifier, or a cable box identifier.

34. The system of claim 31 wherein the numeric transformation uses discrete choice modeling.

35. The system of claim 31 wherein the numeric transformation uses an exponential formula.

36. The system of claim 31 wherein the numeric transformation uses Hayes modeling.