Distributed dialoguing with an agent-oriented software architecture

• US 7,558,819 B1
• Filed: 02/10/2003
• Issued: 07/07/2009
• Est. Priority Date: 02/08/2002
• Status: Active Grant

• Current Assignee: iAnywhere Solutions Incorporated (SAP SE)
• Original Assignee: Sybase Incorporated (SAP SE)
• Inventors: Ito, Junichi, Shiratori, Mitsuru, Funaki, Masahiko, Hodjat, Siamak
• Primary Examiner(s): Hudspeth; David R
• Assistant Examiner(s): Neway; Samuel G
• Application Number:
  US10/361,724
• Time in Patent Office: 2,339 Days
• Field of Search: None
• US Class Current: 709/202
• CPC Class Codes: G06F 16/00 Information retrieval; Database structures therefor; File system structures therefor

Abstract

Roughly described, a natural language interpretation system develops its interpretation in response to user input, and also in dependence upon the interpretation, rather than the claims, made in response to prior input. In policy conditions, an agent can check whether any specified agent contributed to the prior interpretation and adjust its claiming of the present input accordingly. In policy actions, a policy or agent can reference the prior interpretation both for conditioning the policy'"'"'s or agent'"'"'s contribution to the current interpretation, and also for incorporating part or all of the prior interpretation in the policy'"'"'s or agent'"'"'s contribution to the current interpretation. The resulting system thus implements dialoging in a distributed manner as part of the various agents in the interpretation network, and allows much more flexibility to better respond in accordance with users'"'"' expectations.

Claims

1. A method for natural language interpretation, for use with a backend

application and with provided input token sequences, comprising the steps of:

developing a first interpretation in response to a first input token sequence;

representing said first interpretation in a first object; and

developing a second interpretation in response to a second input token sequence, including the step of consulting the first interpretation represented in the first object,

wherein said step of developing a second interpretation comprises the step of forwarding said second input token sequence and said first object toward an agent network,

wherein said agent network comprises a plurality of agents arranged in a network, each of said agents having at least one interpretation policy having a policy condition and a policy action associated with the condition,

wherein said step of developing a second interpretation comprises the step of each of said interpretation policies making a claim on at least part of said second token sequence if the condition of the interpretation policy is satisfied, and executing the associated policy action in response to determination that the claim is at least part of a winning claim,

and wherein a particular one of said interpretation policies includes a policy condition, the satisfaction of which depends on said first interpretation.

2. A method according to claim 1, wherein said steps of developing an

interpretation each comprise the step of developing an interpretation in a distributed manner.

3. A method according to claim 1, wherein said step of developing a second interpretation comprises the step of forwarding said second input token sequence and said first object toward a distributed parser of natural language input.

4. A method according to claim 1, wherein said step of developing a second interpretation policy further comprises the step of a particular one of said agents that includes said particular interpretation policy, referring to a downchain one of said agents in evaluating the policy condition of said particular interpretation policy.

5. A method according to claim 1, wherein said agent network comprises a plurality of agents arranged in a network, each of said agents having at least one interpretation policy having a policy condition and a policy action associated with the condition,

wherein said step of developing a second interpretation comprises the step of each of said interpretation policies making a claim on at least part of said second token sequence if the condition of the interpretation policy is satisfied, and executing the associated policy action in response to determination that the claim is at least part of a winning claim,

and wherein a particular one of said interpretation policies includes a policy action that depends on said first interpretation.

6. A method for natural language interpretation, for use with a network of agents including different first and second agents, for use further with a back-end application and with provided input token sequences, comprising the steps of the network of agents:

developing a first interpretation in response to a first input token sequence, including at least one of said agents contributing to said first interpretation; and

developing a second interpretation in response to a second input token sequence, including said first agent determining whether said first interpretation includes a contribution from said second agent,

wherein said agent network comprises a plurality of agents arranged in a network, each of said agents having at least one interpretation policy having a policy condition and a policy action associated with the condition, said first agent including a particular one of said interpretation policies,

wherein said step of developing a second interpretation comprises the step of each of said interpretation policies making a claim on at least part of said second token sequence if the condition of the interpretation policy is satisfied, and executing the associated policy action in response to determination that the claim is at least part of a winning claim,

and wherein said particular interpretation policy includes a policy condition, the satisfaction of which depends on whether said first interpretation includes a contribution from said second agent.

7. A method according to claim 6, wherein said step of developing a second interpretation further comprises the step of said first agent referring to a downchain one of said agents in evaluating the policy condition of said particular interpretation policy.

8. Natural language interpretation apparatus, for use in developing interpretations for forwarding toward a back-end application, of input token sequences provided to the interpretation apparatus,

comprising data processing hardware programmed to execute a plurality of agents arranged in a network, each of said agents having at least one interpretation policy, each of said interpretation policies having a policy condition and making a claim if the policy condition is satisfied, each of said interpretation policies further contributing to an actuation object representing an interpretation of an input token sequence in response to determination that a claim made by the interpretation policy is at least part of a winning claim,

wherein one of said policy conditions makes reference to a token sequence provided to the agent in response to the input token sequence,

and wherein at least a subject one of said interpretation policies makes reference to a representation of the interpretation of said apparatus in response to a prior input token sequence.

9. Apparatus according to claim 8, wherein at least one of said policy conditions makes reference to a claim made by one of said agents which is downchain of the agent having the policy condition.

10. Apparatus according to claim 8, wherein a particular one of said agents has a particular interpretation policy having a particular policy condition that refers to a downchain one of said agents,

wherein the contribution to said actuation object by said particular interpretation policy includes at least partial delegation to said downchain agent.

11. Apparatus according to claim 8, wherein the contribution to said actuation object by a particular one of said interpretation policies consists of delegation to at least one of said interpretation policies other than said particular interpretation policy.

12. Apparatus according to claim 8, further comprising an interaction entity providing said input token sequences toward said network,

wherein said interaction entity provides toward said network, in conjunction with said input token sequences, a representation of the interpretation of said apparatus in response to said prior input token sequence.

13. Apparatus according to claim 12, wherein said subject interpretation policy is in a subject one of said agents,

wherein said subject agent is operable to, in response to receipt of a token sequence and in conjunction therewith a representation of the interpretation of said apparatus in response to said prior input token sequence, update properties of said subject agent to include said representation,

and wherein said reference made by said subject interpretation policy to a representation of the interpretation of said apparatus in response to a prior input token sequence, includes testing said properties of said subject agent.

14. Apparatus according to claim 12, further comprising an actuation entity, said network providing said actuation object toward said back-end application via said actuation entity,

wherein said actuation entity is operable to forward the interpretation in said actuation object toward said interaction entity, as said representation of the interpretation of said apparatus in response to said prior input token sequence.

15. Apparatus according to claim 14, wherein said actuation entity is operable further to, in at least some circumstances, assert commands toward said back-end application in response to actuation objects received from said network.

16. Apparatus according to claim 8, wherein the reference made by said subject interpretation policy to a representation of the interpretation of said apparatus in response to a prior input token sequence, comprises conditioning said subject interpretation policy on whether any contribution by a referenced one of said agents is included in said interpretation of said apparatus in response to said prior input token sequence.

17. Apparatus according to claim 16, wherein said subject interpretation policy is in a subject one of said agents different from said referenced agent.

18. Apparatus according to claim 8, wherein said network further includes an additional agent.

19. Natural language interpretation apparatus, for use in developing interpretations for forwarding toward a back-end application, of input token sequences provided to the interpretation apparatus,

comprising data processing hardware programmed to execute a plurality of agents arranged in a network and responsive to an input token sequence, at least a subset of said agents each having at least one interpretation policy, each of said interpretation policies having a policy condition and an associated policy action,

each given one of said interpretation policies making a claim in response to receipt by the agent containing the given interpretation policy of a token sequence if the condition of the given interpretation policy is satisfied,

each given one of said interpretation policies further contributing to an actuation object representing an interpretation of an input token sequence in response to determination that a claim made by the given interpretation policy is at least part of a winning claim,

and wherein for at least a subject one of said interpretation policies, when it makes a contribution to an actuation object, the contribution it makes differs depending on the interpretation of said apparatus in response to a prior input token sequence.

20. Apparatus according to claim 19, wherein a particular one of said policy conditions makes reference to a claim made by one of said agents which is downchain of the agent having the particular policy condition.

21. Apparatus according to claim 19, wherein a particular one of said agents has a particular interpretation policy having a particular policy condition that refers to a downchain one of said agents,

wherein the contribution to said actuation object by said particular interpretation policy includes at least partial delegation to said downchain agent.

22. Apparatus according to claim 19, wherein the contribution to said actuation object by a particular one of said interpretation policies consists of delegation to at least one other one of said interpretation policies.

23. Apparatus according to claim 19, further comprising an interaction entity providing said input token sequences toward said network,

wherein said interaction entity provides toward said network, in conjunction with said input token sequences, a representation of the interpretation of said apparatus in response to said prior input token sequence.

24. Apparatus according to claim 23, wherein said function of the interpretation of said apparatus in response to a prior input token sequence is a function of said representation provided by said interaction entity.

25. Apparatus according to claim 23, wherein said subject interpretation policy is in a subject one of said agents,

wherein said subject agent is operable to, in response to receipt of a token sequence and in conjunction therewith a representation of the interpretation of said apparatus in response to said prior input token sequence, update properties of said subject agent to include said representation,

and wherein said function of the interpretation of said apparatus in response to a prior input token sequence is a function of said representation as present in the properties of said subject agent.

26. Apparatus according to claim 23, further comprising an actuation entity, said network providing said actuation object toward said back-end application via said actuation entity,

wherein said actuation entity is operable to forward the interpretation in said actuation object toward said interaction entity, as said representation of the interpretation of said apparatus in response to said prior input token sequence.

27. Apparatus according to claim 26, wherein said actuation entity is further operable to assert commands toward said back-end application in response to actuation objects received from said network.

28. Apparatus according to claim 19, wherein said network further includes an additional agent.