MANAGEMENT OF CONVERSATIONS

• US 20090228264A1
• Filed: 02/03/2009
• Published: 09/10/2009
• Est. Priority Date: 02/11/2003
• Status: Active Grant

• Current Assignee: Microsoft Technology Licensing LLC (Microsoft Corporation)
• Original Assignee: Microsoft Corporation
• Inventors: Williams, David R., Zieman, Yuri, Hill, Jeffrey
• Granted Patent: US 8,000,973 B2
• Time in Patent Office: Days
• US Class Current: 704/9
• CPC Class Codes: G06F 16/3344 using natural language analysis
• CPC Class Codes: G10L 13/027 Concept to speech synthesisers; Generation of natural phrases from machine-based concepts generation of parameters for speech synthesis out of text G10L13/08
• CPC Class Codes: G10L 15/1822 Parsing for meaning understanding
• CPC Class Codes: G10L 15/22 Procedures used during a speech recognition process, e.g. man-machine dialogue
• CPC Class Codes: H04M 3/493 Interactive information services, e.g. directory enquiries ; Arrangements therefor, e.g. interactive voice response [IVR] systems or voice portals
• CPC Class Codes: H04M 3/5166 in combination with interactive voice response systems or voice portals, e.g. as front-ends

Abstract

In one aspect, a communication is received from a user. A concept recognition process is applied to automatically derive a representation of concepts embodied in the communication. The concept representation is used to provide to a human agent information useful in responding to the natural language communication.

Claims

1. A computer-implemented method comprising:

receiving a natural language communication from a user;

utilizing a processing component that is a functional component of the computer to apply a concept recognition process to automatically derive a representation of concepts embodied in the natural language communication;

selecting a collection of information likely to be useful in responding to the natural language communication, wherein selecting comprises identifying the collection of information as corresponding to said representation of concepts;

providing, through an interface that is a functional component of the computer, the collection of information to a human agent;

receiving, through an input device that is a functional component of the computer, an identification of a particular item of information from the collection of information, the particular item being a sub-set of the collection of information, the sub-set being a particular proposed response to the natural language communication; and

delivering the particular proposed response to the user.

2. The method of claim 1, wherein delivering the particular proposed response comprises delivering in a manner that discourages the user from knowing that the response was selected rather than authored by the human agent.

3. The method of claim 1, further comprising initiating a dialog with the user by sending an initial communication to the user, wherein said natural language communication is provided in response to the initial communication.

4. The method of claim 1, wherein the collection of information provided to the human agent includes an audible playback of a recorded version of at least a portion of the natural language communication received from the user.

5. The method of claim 4, wherein the audible playback is compressed in time relative to the natural language communication received from the user.

6. The method of claim 1, wherein the collection of information provided to the human agent includes a display of text indicative of at least a portion of the natural language communication received from the user.

7. The method of claim 1, wherein the collection of information provided to the human agent includes a record of additional communication from the user received prior to said receiving of the natural language communication.

8. The method of claim 7, wherein said record of additional communication from the user is a record of a number of additional communications from the user received prior to said receiving of the natural language communication, and wherein said number is a value determined at least in part based on which concepts are included in the representation of concepts.

9. The method of claim 1, wherein the natural language communication is received by way of a first mode of communication and the particular proposed response is delivered by way of a second mode of communication, the first mode of communication being different than the second mode of communication.

10. The method of claim 1, further comprising:

receiving, from the human agent, an edit to the particular proposed response; and

applying the edit to the particular proposed response before the particular proposed response is delivered to the user, wherein applying the edit comprises altering content of the particular proposed response.

11. The method of claim 1, wherein selection a collection of information comprises applying a matching process so as to automatically determine a cluster of similar communications to which the representation of concepts corresponds.

12. A computer-implemented method comprising:

receiving a natural language communication from a user;

determining a degree of involvement of a human agent necessary to respond to the natural language communication received from the user;

selecting one of a plurality of conversation management modes, the selection being made at least partially contingent upon the determined degree involvement, wherein selecting is a function performed by a computer processor that is a function component of the computer, and wherein the computer processor facilitates an actual execution of the selected conversation management mode, and wherein;

a first mode of the plurality of conversation management modes comprises providing a response to the user in an automated manner without interaction by the human agent;

a second mode of the plurality of conversation management modes comprises providing, to the user, a response at least initially selected by the human agent from a closed set of possible responses; and

a third mode of the plurality of conversation management modes comprises initiating direct communication between the human agent and the user.

13. The method of claim 12, wherein the first second and third mode of the plurality of conversation management modes are all pre-configured for selection by the computer processor and execution regardless of which particular mode is actually selected as part of said step of selecting at least one mode.

14. The method of claim 12, wherein the computer processor is configured to, based on an estimate of how long it would take for the human agent to respond, bias said selection of the conversation management mode toward one of either the first, or third mode.

15. The method of claim 12, wherein determining a degree of involvement includes determining an estimate of how long it would take for the human agent to respond, said selection being made at least partially contingent upon the estimate.

16. The method of claim 12, wherein determining a degree of involvement includes determining a level of confidence representing a likelihood that the closed set of possible responses is a desirable response to the natural language communication received from the user.

17. A computer-implemented method comprising:

receiving a communication from a user;

providing the user with a response to the communication;

wherein the response is selected by a human agent from a set of candidate responses automatically generated by a computer processor that is a functional component of the computer, wherein the set of candidate responses is derived by the processor by automatically identifying a concept associated with the communication, wherein the set of candidate responses is limited to responses that are identified, by the computer processor, as being consistent with the identified concept; and

delivering the response to the user in a manner that discourages the user from knowing that the response was selected by the human agent from the set of candidate responses.

18. The method of claim 17, wherein the communication is received by way of a first mode of communication and the response is delivered by way of a second mode of communication, the first mode of communication being different than the second mode of communication.

19. The method of claim 18, wherein only one of the first and second modes of communications is a speech mode of communication.

20. The method of claim 18, wherein only one of the first and second modes of communications is a text-based mode of communication.