Micro-dynamic simulation of electronic cash transactions

• US 5,949,045 A
• Filed: 07/03/1997
• Issued: 09/07/1999
• Est. Priority Date: 07/03/1997
• Status: Expired due to Term

• Current Assignee: AT&T Corporation (AT&T, Inc.)
• Original Assignee: AT&T Corporation (AT&T, Inc.)
• Inventors: Napiorkowski, Gregory, Ezawa, Kazuo J.
• Primary Examiner(s): Le, Thien Minh
• Assistant Examiner(s): FUREMAN, JARED
• Application Number:
  US08/888,024
• Time in Patent Office: 796 Days
• Field of Search: 235/375, 235/379, 235/380, 235/492, 705/16, 705/17, 705/39, 705/2, 902/22, 902/26
• US Class Current: 235/379
• CPC Class Codes: G06Q 20/20 Point-of-sale [POS] network systems
• CPC Class Codes: G06Q 30/06 Buying, selling or leasing transactions
• CPC Class Codes: G06Q 40/08 Insurance

Abstract

A technique for representing behavioral patterns indicating fraudulent electronic cash transactions through micro-dynamic simulation of an electronic cash transaction system implementable using smart cards with embedded microcomputers. The electronic cash transaction system is stochastically modeled by four categories of agents corresponding to the participants in an electronic cash transaction economy. Some categories of agents are further subdivided into segments that have different characteristics, including the circle of other agents with whom they regularly conduct electronic cash transactions. Agents with counterfeit or fraudulent smart cards may form a segment in the simulation to allow the effects of the counterfeit or fraudulent electronic cash transactions in the complete system to be observed.

Claims

1. A method of modeling an electronic cash transaction system by computerized simulation comprising the steps of:

defining the simulation parameters;

defining the characteristics of at least one creator agent of the electronic cash transaction system;

defining the characteristics of smart card issuer agents of the electronic cash transaction system;

defining the characteristics of merchant agents of the electronic cash transaction system;

defining the characteristics of consumer agents of the electronic cash transaction system; and

running a computer simulation in accordance with said parameters and said characteristics and thereby generating data indicative of electronic cash transactions conducted by said agents.

2. The method of modeling an electronic cash transaction system by computerized simulation of claim 1 further comprising the step of defining circles indicative of the agents with which an agent conducts the majority of said agent'"'"'s electronic cash transactions.

3. The method of modeling an electronic cash transaction system by computerized simulation of claim 2 wherein said step of defining circles comprises the steps of:

defining the maximum possible number of agents within said agent'"'"'s circle;

defining the minimum possible number of agents within said agent'"'"'s circle; and

defining the probability of a transaction by said agent being conducted with an agent in said agent'"'"'s circle.

4. The method of modeling an electronic cash transaction system by computerized simulation of claim 1 wherein said step of defining the characteristics of card issuer agents of the electronic cash transaction system comprises the steps of:

defining the purse characteristics of said card issuer agents;

defining the categories of agents with which said card issuer agents transact; and

defining the population characteristics of said card issuer agents.

5. The method of modeling an electronic cash transaction system by computerized simulation of claim 1 wherein said step of defining the characteristics of merchant agents of the electronic cash transaction system comprises the steps of:

defining the purse characteristics of said merchant agents;

defining the categories of agents with which said merchant agents transact; and

defining the population characteristics of said merchant agents.

6. The method of modeling an electronic cash transaction system by computerized simulation of claim 1 wherein said step of defining the characteristics of consumer agents of the electronic cash transaction system comprises the steps of:

defining the purse characteristics of said consumer agents;

defining the transaction characteristics of said consumer agents;

defining the categories of agents with which said consumer agents transact; and

defining the population characteristics of said consumer agents.

7. The method of modeling a smart card based electronic cash transaction system by computerized simulation of claim 6 wherein said step of defining the characteristics of consumer agents of the electronic cash transaction system further comprises a step of defining conditions under which said consumer agent is prohibited from conducting further transactions.

8. A method of modeling transactions by stochastic simulation comprising the steps of:

defining a plurality of agents conducting transactions;

defining the transaction characteristics for each of said plurality of agents using probability functions;

defining a circle of said agents with which each of said plurality of agents has a probability of conducting a transaction provided a transaction is conducted; and

simulating transactions conducted by said plurality of agents in accordance with said defined transaction characteristics.

9. The method of modeling transactions by stochastic simulation of claim 8 wherein said step of defining a circle of said agents with which each of said plurality of agents has a probability of conducting a transaction when each of said plurality of agents conducts a transaction comprises:

defining probabilistically the number of agents inside of the circle;

defining probabilistically the number of transactions that are conducted with an agent inside of the circle; and

defining probabilistically the number of transactions that are conducted with an agent outside of the circle.

10. The method of modeling transactions by stochastic simulation of claim 8 wherein said step of defining a circle of said agents with which each of said plurality of agents has a probability of conducting a transaction when each of said plurality of agents conducts a transaction comprises:

defining probabilistically the number of agents inside the circle;

defining probabilistically the proportion of transactions that are conducted with an agent inside the circle; and

defining probabilistically the proportion of transactions that are conducted with agents that are outside the circle.

11. The method of-modeling transactions by stochastic simulation of claim 8 wherein said step of defining a circle of said agents with which each of said plurality of agents has a probability of conducting a transaction when each of said plurality of agents conducts a transaction further comprises:

defining the maximum number of said plurality of agents with which each of said plurality of agents has a probability of conducting a transaction when each of said plurality of agents conducts a transaction;

defining the minimum number of said plurality of agents with which each of said plurality of agents has a probability of conducting a transaction when each of said plurality of agents conducts a transaction; and

defining the probability of a transaction being conducted with a member of said circle of said agents when a transaction is conducted.

12. The method of modeling transactions by stochastic simulation of claim 8 further comprising the steps of:

associating each of said plurality of agents with a category of agents wherein said associated category determines with which of said plurality of agents each of said plurality of agents may conduct transactions; and

segmenting each of said category of agents into at least one segment.

13. The method of modeling transactions by stochastic simulation of claim 12 wherein said step of defining a circle of said agents with which each of said plurality of agents has a probability of conducting a transaction when each of said plurality of agents conducts a transaction further comprises defining the segment with which each of said agents within said circle is associated.

14. A method of providing input data to machine learning software by computerized simulation of an electronic cash transaction system comprising the steps of:

defining the simulation parameters;

defining the characteristics of at least one creator agent of the electronic cash transaction system;

defining the characteristics of smart card issuer agents of the electronic cash transaction system;

defining the characteristics of merchant agents of the electronic cash transaction system;

defining the characteristics of consumer agents of the electronic cash transaction system; and

running a computer simulation in accordance with said parameters and said characteristics and thereby generating data indicative of electronic cash transactions conducted by said agents;

providing the results of said computer simulation to machine learning software as input data.

15. The method of providing input data to machine learning software by computerized simulation of an electronic cash transaction system of claim 14 further comprising the step of defining circles indicative of the agents with which an agent conducts the majority of said agent'"'"'s electronic cash transactions.

16. The method of providing input data to machine learning software by computerized simulation of an electronic cash transaction system of claim 15 wherein said step of defining circles comprises the steps of:

defining the maximum possible number of agents within said agent'"'"'s circle;

defining the minimum possible number of agents within said agent'"'"'s circle; and

defining the probability of a transaction by said agent being conducted with an agent in said agent'"'"'s circle.

17. The method of providing input data to machine learning software by computerized simulation of an electronic cash transaction system of claim 16 wherein said step of defining the characteristics of card issuer agents of the electronic cash transaction system comprises the steps of:

defining the purse characteristics of said card issuer agents;

defining the categories of agents with which said card issuer agents transact; and

defining the population characteristics of said card issuer agents.

18. The method of providing input data to machine learning software by computerized simulation of an electronic cash transaction system of claim 14 wherein said step of defining the characteristics of merchant agents of the electronic cash transaction system comprises the steps of:

defining the purse characteristics of said merchant agents;

defining the categories of agents with which said merchant agents transact; and

defining the population characteristics of said merchant agents.

19. The method of providing input data to machine learning software by computerized simulation of an electronic cash transaction system of claim 14 wherein said step of defining the characteristics of consumer agents of the electronic cash transaction system comprises the steps of:

defining the purse characteristics of said consumer agents;

defining the transaction characteristics of said consumer agents;

defining the categories of agents with which said consumer agents transact; and

defining the population characteristics of said consumer agents.

20. The method of providing input data to machine learning software by computerized simulation of an electronic cash transaction system of claim 19 wherein said step of defining the characteristics of consumer agents of the electronic cash transaction system further comprises a step of defining conditions under which said consumer agent is prohibited from conducting further transactions.