Intelligent agent with negotiation capability and method of negotiation therewith

• US 6,401,080 B1
• Filed: 03/21/1997
• Issued: 06/04/2002
• Est. Priority Date: 03/21/1997
• Status: Expired due to Fees

• Current Assignee: International Business Machines Corporation
• Original Assignee: International Business Machines Corporation
• Inventors: Delp, Helen Roxlo, Cragun, Brian John, Bigus, Joseph Phillip
• Primary Examiner(s): Hafiz, Tariq R.
• Assistant Examiner(s): KALINOWSKI, ALEXANDER G
• Application Number:
  US08/821,935
• Time in Patent Office: 1,901 Days
• Field of Search: 705/37, 705/1, 705/35, 705/26, 705/44, 705/400
• US Class Current: 705/37
• CPC Class Codes: G06Q 20/3821 Electronic credentials
• CPC Class Codes: G06Q 30/06 Buying, selling or leasing transactions
• CPC Class Codes: G06Q 40/00 Finance; Insurance; Tax strategies; Processing of corporate or income taxes
• CPC Class Codes: G06Q 40/04 Trading; Exchange, e.g. stocks, commodities, derivatives or currency exchange

Abstract

An intelligent agent and method of negotiating herewith incorporate a number of features, used alone or in combination, to enhance the productivity, security, efficiency and responsiveness of the agent in negotiations with other parties. One feature incorporates randomization of one or more aspects of an agent'"'"'s behavior to disguise its negotiation strategy from other negotiating parties and thereby prevent such parties from gaining a negotiating advantage at the expense of the agent. Another features incorporates limiting unproductive negotiations by constraining one or more aspects of an agent'"'"'s behavior based upon the behavior of a negotiating party and/or the duration of the transaction, and thereby making it more likely that unproductive negotiations will be terminated. An additional feature incorporates dynamic value determination to determine the desired value of a desired transaction by weighting and normalizing estimated values retrieved from a plurality of information sources. Moreover, a further features incorporates dynamic value determination which weights and normalizes the values of related transactions based upon the proximity of the related and desired transactions.

Claims

1. A method of conducting an electronic transaction with an intelligent agent, the method comprising the steps of:

(a) generating an offer to enter into a transaction;

(b) waiting for a response from a negotiating party;

(c) upon receiving a response, determining whether to complete the transaction; and

(d) disguising a negotiation strategy from the negotiating party by randomizing a characteristic of at least one of the generating, waiting and determining steps.

2. The method of claim 1, further comprising the step of limiting unproductive negotiations by constraining a characteristic of at least one of the generating, waiting and determining steps based upon at least one of a behavior of the negotiating party and a duration of the transaction.

3. The method of claim 1, wherein the step of determining whether to complete the transaction includes the steps of:

(a) completing the transaction if an asked price from one of the offer and the response is less than or equal to a bid price from the other of the offer and the response;

(b) determining an accept probability value based upon the duration of the transaction and the proximity of the offer and response, the accept probability value dividing probability range into accept and reject portions;

(c) selecting a random number within the probability range; and

(d) completing the transaction if the random number falls within the accept portion of the probability range.

4. The method of claim 1, further comprising the steps of monitoring transactions with a high pass filter that calculates a market slope to detect volatile market conditions, and adjusting the negotiation strategy upon detection of volatile market conditions.

5. The method of claim 4, further comprising the step of locking out at least one of stop loss and stop gain offers in response to an inversion of the market slope calculated by the high pass filter.

6. The method of claim 1, wherein the offer generating step includes the steps of determining a value for a desired transaction and calculating a range of acceptable offer prices from the value, and wherein the disguising step includes the step of selecting for the offer a random price within the range of acceptable offer prices.

7. The method of claim 6, wherein the value determining step includes the steps of:

(a) retrieving a plurality of related transactions that are related to the desired transaction, each related transaction having a value associated therewith;

(b) for each related transaction, weighting the value of the related transaction based upon a proximity between the related and desired transactions to obtain a weighted value; and

(c) normalizing the weighted values to generate the value for the desired transaction therefrom.

8. The method of claim 6, wherein the value determining step includes the steps of:

(a) generating a plurality of estimated values from a plurality of information sources;

(b) weighting the plurality of estimated values based upon a predetermined criteria to generate a plurality of weighted estimated values; and

(c) normalizing the plurality of weighted estimated values to generate the value for the desired transaction thereform.

9. The method of claim 6, wherein the offer generating step further comprises the steps of:

(a) storing previous asked and bid prices; and

(b) adjusting the previous asked and bid prices in response to a change in the value of the desired transaction.

10. The method of claim 6, wherein the agent operates as a selling agent, and wherein the calculating the range of acceptable prices step includes the steps of:

(a) selecting a minimum offer price from the maximum of (1) a previous asked price and (2) the value of the desired transaction plus a required profit margin; and

(b) selecting a maximum offer price from the minimum of (1) a previous bid price and (2) the value of the desired transaction plus the required profit and a negotiating margin.

11. The method of claim 6, wherein the agent operates as a buying agent, and wherein the calculating the range of acceptable prices step includes the steps of:

(a) selecting a minimum offer price from the maximum of (1) a previous bid price and (2) the value of the desired transaction less a required profit margin and less a negotiating margin; and

(b) selecting a maximum offer price from the minimum of (1) a previous asked price and (2) the value of the desired transaction less the required profit.

12. The method of claim 6, wherein the offer generating step further comprises the steps of:

(a) detecting a real price for the negotiating party; and

(b) constraining the range of acceptable offer prices using the real price.

13. The method of claim 1, wherein the waiting step further includes the step of selecting a random wait time between maximum and minimum offer duration times.

14. The method of claim 13, wherein the waiting step further includes the step of calculating a wait probability value based upon the duration of the transaction and a last offer received from the negotiating party.

15. The method of claim 14, wherein the waiting step further includes the step of generating a probability distribution from the wait probability value, and wherein the step of selecting a random wait time selects a random point in the probability distribution when selecting a wait time between the maximum and minimum offer duration times.

16. The method of claim 1, further comprising the step of determining whether to make a counteroffer, wherein the disguising step includes the step of randomizing a characteristic of the determining whether to make a counteroffer step.

17. The method of claim 11, wherein the determining whether to make a counteroffer step includes the steps of:

(a) determining a counteroffer probability value based upon the duration of the transaction and the proximity of the offer and response, the counteroffer probability value dividing a probability range into counteroffer and no counteroffer portions;

(b) selecting a random number within the probability range; and

(c) making a counteroffer if the random number falls within the counteroffer portion of the probability range.

18. The method of claim 17, further comprising the step of waiting for a randomized time period prior to making a counteroffer.

19. A method of conducting an electronic transaction with an intelligent agent, the method comprising the steps of:

(a) generating an offer to enter into a transaction;

(b) waiting for a response from a negotiating party;

(c) upon receiving a response, determining whether to complete the transaction;

(d) limiting unproductive negotiations by constraining a characteristic of at least one of the generating, waiting and determining steps based upon at least one of a behavior of the negotiating party and a duration of the transaction; and

(e) disguising a negotiation strategy from the negotiating party by randomizing a characteristic of at least one of the generating, waiting and determining steps.

20. The method of claim 19, wherein the offer generating step of determining a value for a desired transaction, and wherein the constraining step includes the step of calculating a range of acceptable offer prices from the value.

21. The method of claim 20, wherein the value determining step includes the steps of:

(a) retrieving a plurality of related transactions that are related to the desired transaction, each related transaction having a value associated therewith;

(b) for each related transaction, weighting the value of the related transaction based upon a proximity between the related and desired transactions to obtain a weighted value; and

(c) normalizing the weighted values to generate the value for the desired transaction therefrom.

22. The method of claim 20, wherein the value determining step includes the steps of:

(a) generating a plurality of estimated values from a plurality of information sources;

(b) weighting the plurality of estimated values based upon a predetermined criteria to generate a plurality of weighted estimated values; and

(c) normalizing the plurality of weighted estimated values to generate the value for the desired transaction therefrom.

23. The method of claim 20, wherein the agent operates as a selling agent, and wherein the calculating the range of acceptable prices step includes the steps of:

(a) selecting a minimum offer price from the maximum of (1) a previous asked price and (2) the value of the desired transaction plus a required profit margin; and

(b) selecting a maximum offer price from the minimum of (1) a previous bid price and (2) the value of the desired transaction plus the required profit and a negotiating margin.

24. The method of claim 20, wherein the agent operates as a buying agent, and wherein the calculating the range of acceptable prices step includes the steps of:

(a) selecting a minimum offer price from the maximum of (1) a previous bid price and (2) the value of the desired transaction less a required profit margin and less a negotiating margin; and

(b) selecting a maximum offer price from the minimum of (1) a previous asked price and (2) the value of the desired transaction less the required profit.

25. The method of claim 20, wherein the offer generating step comprises the step of detecting a real price for the negotiating party, and wherein the constraining step includes the step of constraining the range of acceptable offer prices using the real price.

26. An apparatus for conducting an electronic transaction, the apparatus including an intelligent agent computer program executed by the apparatus, the intelligent agent including an agent negotiation module configured to perform the method steps of:

(a) generating an offer to enter into a transaction;

(b) waiting for a response from a negotiating party;

(c) upon receiving a response, determining whether to complete the transaction; and

(d) disguising a negotiation strategy for the intelligent agent from the negotiating party by randomizing a characteristic of at least one of the generating, waiting and determining steps.

27. The apparatus of claim 26, wherein the agent negotiation module is further configured to execute the step of limiting unproductive negotiations by constraining a characteristic of at least one of the generating, waiting and determining steps based upon at least one of a behavior of the negotiating party and a duration of the transaction.

28. The apparatus of claim 26, wherein the agent negotiation module is configured to select a random wait time between maximum and minimum offer duration times.

29. The apparatus of claim 26, wherein the agent negotiation module is further configured to:

(a) complete the transaction if an asked price from one of the offer and the response is less than or equal to a bid price from the other of the offer and the response;

(b) determine an accept probability value based upon the duration of the transaction and the proximity of the offer and response, the accept probability value dividing a probability range into accept and reject portions;

(c) select a random number within the probability range; and

(d) complete the transaction if the random number falls within the accept portion of the probability range.

30. The apparatus of claim 26, wherein the intelligent agent further comprises a high pass filter that calculates a market slope to detect volatile market conditions, and adjusts the negotiation strategy upon detection of volatile market conditions.

31. The apparatus of claim 26, wherein the intelligent agent further includes a value determination module configured to determine a value for a desired transaction, and wherein the agent negotiation module is further configured to calculate a range of acceptable offer prices from the value.

32. The apparatus of claim 31, wherein the intelligent agent operates as a selling agent, and wherein the agent negotiation module is configured to:

(a) select a minimum offer price from the maximum of (1) a previous asked price and (2) the value of the desired transaction plus a required profit margin; and

(b) select a maximum offer price from the minimum of (1) a previous bid price and (2) the value of the desired transaction plus the required profit and a negotiating margin.

33. The apparatus of claim 31, wherein the intelligent agent operates as a buying agent, and wherein the agent negotiation module is configured to:

(a) select a minimum offer price from the maximum of (1) a previous bid price and (2) the value of the desired transaction less a required profit margin and less a negotiating margin; and

(b) select a maximum offer price from the minimum of (1) a previous asked price and (2) the value of the desired transaction less the required profit.

34. A program product comprising:

(a) a program configured to perform a method of conducting an electronic transaction, the method comprising the steps of;

(1) generating an offer to enter into a transaction;

(2) waiting for a response from a negotiating party;

(3) upon receiving a response, determining whether to complete the transaction; and

(4) disguising a negotiation strategy from the negotiating party by randomizing a characteristic of at least one of the generating, waiting and determining steps; and

(b) a signal bearing media bearing the program.

35. The program product of claim 34, wherein the signal bearing media is transmission type media.

36. The program product of claim 34, wherein the signal bearing media is recordable media.

37. The program product of claim 34, wherein the program is an intelligent agent.

38. A method of conducting an electronic transaction with an intelligent agent, the method comprising the steps of:

(a) generating an offer to enter into a transaction;

(b) waiting for a response from a negotiating party;

(c) upon receiving a response, determining whether to complete the transaction; and

(d) limiting unproductive negotiations by constraining a characteristic of at least one of the generating, waiting and determining steps based upon at least one of a behavior of the negotiating party and a duration of the transaction and a last offer received from the negotiating party, and wherein the constraining step further includes the steps of;

(i) generating a probability distribution from the wait probability value; and

(ii) selecting a random point in the probability distribution to generate an offer duration for the waiting step.

39. A method of conducting an electronic transaction with an intelligent agent, the method comprising the steps of:

(a) generating an offer to enter into a transaction;

(b) waiting for a response from a negotiating party;

(c) upon receiving a response, determining whether to complete the transaction; and

(d) limiting unproductive negotiations by constraining a characteristic of at least one of the generating, waiting and determining steps based upon at least one of a behavior of the negotiating party and a duration of the transaction and a last offer received from the negotiating party, and wherein the constraining step further includes the steps of;

(i) determining an accept probability value based upon the duration of the transaction and the proximity of the offer and response, the accept probability value dividing a probability range into accept and reject portions;

(ii) selecting a random number within the probability range; and

(iii) completing the transaction if the random number falls within the accept portion of the probability range.

40. A method of conducting an electronic transaction with an intelligent agent, the method comprising the steps of:

(a) generating an offer to enter into a transaction;

(b) waiting for a response from a negotiating party;

(c) upon receiving a response, determining whether to complete the transaction;

(d) limiting unproductive negotiations by constraining a characteristic of at least one of the generating, waiting and determining steps based upon at least one of a behavior of the negotiating party and a duration of the transaction, wherein the constraining step includes the step of calculating a wait probability value based upon the duration of the transaction and a last offer received from the negotiating party, and

(e) determining whether to make a counteroffer, including the steps of;

(i) determining a counteroffer probability value based upon the duration of the transaction and the proximity of the offer and response, the counteroffer probability value dividing a probability range into counteroffer and no counteroffer portions;

(ii) selecting a random number within the probability range; and

(iii) making a counteroffer if the random number falls within the counteroffer portion of the probability range.

41. An apparatus for conducting an electronic transaction, the apparatus including an intelligent agent computer program executed by the apparatus, the intelligent agent including an agent negotiation module configured to perform the method steps of:

(a) generating an offer to enter into a transaction;

(b) waiting for a response from a negotiating party;

(c) upon receiving a response, determining whether to complete the transaction; and

(d) limiting unproductive negotiations by constraining a characteristic of at least one of the generating, waiting and determining steps based upon at least one of a behavior of the negotiating party and a duration of the transaction;

wherein the agent negotiation module is further configured to disguise a negotiation strategy from the negotiating party by randomizing a characteristic of at least one of the generating, waiting and determining steps.

42. The apparatus of claim 41, wherein the agent negotiation module is further configured to:

(a) determine an accept probability value based upon the duration of the transaction and the proximity of the offer and response, the accept probability value dividing a probability range into accept and reject portions;

(b) select a random number within the probability range; and

(c) complete the transaction if the random number falls within the accept portion of the probability range.

43. The apparatus of claim 41, wherein the agent negotiation module is further configured to:

(a) determine a counteroffer probability value based upon the duration of the transaction and the proximity of the offer and response, the counteroffer probability value dividing a probability range into counteroffer and no counteroffer portions;

(b) select a random number within the probability range; and

(c) make a counteroffer if the random number falls within the counteroffer portion of the probability range.

44. An apparatus for conducting an electronic transaction, the apparatus including an intelligent agent computer program executed by the apparatus, the intelligent agent including an agent negotiation module configured to perform the method steps of:

(a) generating an offer to enter into a transaction;

(b) waiting for a response from a negotiating party;

(c) upon receiving a response, determining whether to complete the transaction; and

(d) limiting unproductive negotiations by constraining a characteristic of the generating step based upon a behavior of the negotiating party;

wherein the intelligent agent further includes a value determination module configured to determine a value for a desired transaction, and wherein the agent negotiation module is further configured to calculate a range of acceptable offer prices from the value.

45. The apparatus of claim 44, wherein the intelligent agent operates as a selling agent, and wherein the agent negotiation module is configured to:

(a) select a minimum offer price from the maximum of (1) a previous asked price and (2) the value of the desired transaction plus a required profit margin; and

(b) select a maximum offer price from the minimum of (1) a previous bid price and (2) the value of the desired transaction plus the required profit and a negotiating margin.

46. The apparatus of claim 45, wherein the intelligent agent operates as a buying agent, and wherein the agent negotiation module is configured to:

(a) select a minimum offer price from the maximum of (1) a previous bid price and (2) the value of the desired transaction less a profit margin and less a negotiating margin; and

(b) select a maximum offer price from the minimum of (1) a previous asked price and (2) the value of the desired transaction less the required profit.

47. A program product comprising:

(a) a program configured to perform a method of conducting an electronic transaction, the method comprising the steps of;

(1) generating an offer to enter into a transaction;

(2) waiting for a response from a negotiating party;

(3) upon receiving a response, determining whether to complete the transaction;

(4) limiting unproductive negotiations by constraining a characteristic of at least one of the generating, waiting and determining steps based upon at least one of a behavior of the negotiating party and a duration of the transaction; and

(5) disguising a negotiation strategy from the negotiating party by randomizing a characteristic of at least one of the generating, waiting and determining steps; and

(b) a signal bearing media bearing the program.

48. The program product of claim 47, wherein the signal bearing media is transmission type media.

49. The program product of claim 47, wherein the signal bearing media is recordable media.

50. The program product of claim 47, wherein the program is an intelligent agent.

51. A method of conducting an electronic transaction with an intelligent agent, the method comprising the steps of:

(a) generating an offer to enter into a transaction, wherein the offer generating steps includes the steps of;

(1) determining a value for a desired transaction;

(2) calculating a range of acceptable offer prices from the value;

(3) storing previous asked and bid prices; and

(4) adjusting the previous asked and bid prices in response to a change in the value of the desired transaction;

(b) waiting for a response from a negotiating party;

(c) upon receiving a response, determining whether to complete the transaction; and

(d) disguising a negotiation strategy from the negotiating party by randomizing a characteristic of at least one of the generating, waiting and determining steps, wherein the disguising step includes the step of selecting for the offer a random price within the range of acceptable offer prices.

52. A method of conducting an electronic transaction with an intelligent agent, the method comprising the steps of:

(a) generating an offer to enter into a transaction, wherein the offer generating step includes the steps of;

(1) determining a value for a desired transaction;

(2) calculating a range of acceptable offer prices from the value;

(3) detecting a real price for a negotiating party; and

(4) constraining the range of acceptable offer prices using the real price.

(b) waiting for a response from the negotiating party;

(c) upon receiving a response, determining whether to complete the transaction; and

(d) disguising a negotiation strategy from the negotiating party by randomizing a characteristic of at least one of the generating, waiting and determining steps, wherein the disguising step includes the step of selecting for the offer a random price within the range of acceptable offer prices.