Intelligent agent with negotiation capability and method of negotiation therewith

• US 7,908,225 B1
• Filed: 11/02/1999
• Issued: 03/15/2011
• Est. Priority Date: 03/21/1997
• Status: Expired due to Fees

• Current Assignee: International Business Machines Corporation
• Original Assignee: International Business Machines Corporation
• Inventors: Bigus, Joseph Phillip, Delp, Helen Roxlo, Cragun, Brian John
• Primary Examiner(s): Fischer; Andrew J.
• Assistant Examiner(s): Coppola; Jacob C.
• Application Number:
  US09/431,833
• Time in Patent Office: 4,151 Days
• Field of Search: 705/80, 705/76, 706/46, 706/16, 706/10, 376/264
• US Class Current: 705/76
• 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 therewith 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 identifying an unknown party interacting with an intelligent agent, the method comprising, in a computer that includes at least one processor, executing a program to perform the steps of:

determining a plurality of attributes related to the unknown party, wherein the unknown party is a party other than a client that has delegated at least one task to the intelligent agent;

comparing the plurality of attributes for the unknown party with attributes related to a plurality of known parties; and

identifying the unknown party as the known party having attributes that most closely match those of the unknown party;

wherein the unknown party is an intelligent agent configured to conduct electronic transactions, wherein the plurality of attributes is selected from the group consisting of an agent name, a client name, a bank name, a bank account number, a credit card number, a homebase location, an agent program name, or a location or name of a source with which the unknown party communicates, wherein each of the plurality of attributes has a weighting factor associated therewith, wherein the comparing step calculates an accumulated weighting factor for each known party by summing the weighting factors of the attributes of the known party which match those of the unknown party, and wherein the identifying step identifies the unknown party as the known party with the largest accumulated weighting factor.

2. The method of claim 1, wherein the comparing step includes the step of accessing a database including a plurality of records, each record associated with a known party and including the plurality of attributes related to the known party.

3. The method of claim 1, wherein the determining step includes the step of scanning program code for the unknown party to determine attributes thereof.

4. An apparatus for identifying an unknown party interacting with an intelligent agent, comprising:

at least one processor;

a database including a plurality of records, each record associated with a known party and including a plurality of attributes related to the known party; and

an identification module, coupled to the database, the identification module comprising instructions configured upon execution by the at least one processor to compare a plurality of attributes for the unknown party with those of each known party and to identify the unknown party as the known party having attributes that most closely match those of the unknown party, wherein the unknown party is a party other than a client that has delegated at least one task to the intelligent agent;

wherein the unknown party is an intelligent agent configured to conduct electronic transactions, wherein the plurality of attributes are selected from the group consisting of an agent name, a client name, a bank name, a bank account number, a credit card number, a homebase location, an agent program name, or a location or name of a source with which the unknown party communicates, wherein each of the plurality of attributes has a weighting factor associated therewith, and wherein the identification module is configured to calculate an accumulated weighting factor for each known party by summing the weighting factors of the attributes of the known party which match those of the unknown party, and to identify the unknown party as the known party with the largest accumulated weighting factor.

6. The apparatus of claim 4, wherein the identification module is configured to scan program code for the unknown party to determine attributes thereof.

5. A program product comprising:

a recordable computer readable medium; and

a program stored on the recordable computer readable medium and configured upon execution to perform a method of identifying an unknown party interacting with an intelligent agent, the method comprising the steps of;

determining a plurality of attributes related to the unknown party, wherein the unknown party is a party other than a client that has delegated at least one task to the intelligent agent;

comparing the plurality of attributes for the unknown party with attributes related to a plurality of known parties; and

identifying the unknown party as the known party having attributes that most closely match those of the unknown party;

wherein the unknown party is an intelligent agent configured to conduct electronic transactions, wherein the plurality of attributes are selected from the group consisting of an agent name, a client name, a bank name, a bank account number, a credit card number, a homebase location, an agent program name, or a location or name of a source with which the unknown party communicates, wherein each of the plurality of attributes has a weighting factor associated therewith, and wherein the program is configured to calculate an accumulated weighting factor for each known party by summing the weighting factors of the attributes of the known party which match those of the unknown party, and to identify the unknown party as the known party with the largest accumulated weighting factor.

7. The program product of claim 5, wherein the program is configured to access a database including a plurality of records, each record associated with a known party and including the plurality of attributes related to the known party.

8. The program product of claim 5, wherein the program is configured to scan program code for the unknown party to determine attributes thereof.

9. A method of identifying an unknown party interacting with a first intelligent agent, the method comprising, in a computer that includes at least one processor, executing a program to perform the steps of:

determining a plurality of attributes related to the unknown party, wherein the unknown party is a second intelligent agent interacting with the first intelligent agent;

comparing the plurality of attributes for the unknown party with attributes related to a plurality of known parties; and

identifying the unknown party as the known party having attributes that most closely match those of the unknown party;

wherein the unknown party is configured to conduct electronic transactions, wherein the plurality of attributes is selected from the group consisting of an agent name, a client name, a bank name, a bank account number, a credit card number, a homebase location, an agent program name, or a location or name of a source with which the unknown party communicates, wherein each of the plurality of attributes has a weighting factor associated therewith, wherein the comparing step calculates an accumulated weighting factor for each known party by summing the weighting factors of the attributes of the known party which match those of the unknown party, and wherein the identifying step identifies the unknown party as the known party with the largest accumulated weighting factor.

10. A method of identifying an unknown party interacting with an intelligent agent, the method comprising, in a computer that includes at least one processor, executing a program to perform the steps of:

determining a plurality of attributes related to the unknown party;

comparing the plurality of attributes for the unknown party with attributes related to a plurality of known parties;

identifying the unknown party as the known party having attributes that most closely match those of the unknown party; and

controlling a behavior of the intelligent agent when interacting with the unknown party based upon the identification of the unknown party;

wherein the unknown party is an intelligent agent configured to conduct electronic transactions, wherein the plurality of attributes is selected from the group consisting of an agent name, a client name, a bank name, a bank account number, a credit card number, a homebase location, an agent program name, or a location or name of a source with which the unknown party communicates, wherein each of the plurality of attributes has a weighting factor associated therewith, wherein the comparing step calculates an accumulated weighting factor for each known party by summing the weighting factors of the attributes of the known party which match those of the unknown party, and wherein the identifying step identifies the unknown party as the known party with the largest accumulated weighting factor.

11. The method of claim 10, wherein controlling the behavior of the intelligent agent includes controlling a negotiation strategy used by the intelligent agent when conducting an electronic transaction with the unknown party.

12. The method of claim 10, wherein identifying the unknown party includes identifying the unknown party as being untrustworthy, and wherein controlling the behavior of the intelligent agent includes modifying the behavior of the intelligent agent to account for increased risk posed by the unknown party and continuing to interact with the unknown party using the modified behavior.