PARAMETERIZED STATISTICAL INTERACTION POLICIES

• US 20080313116A1
• Filed: 06/13/2007
• Published: 12/18/2008
• Est. Priority Date: 06/13/2007
• Status: Active Grant

• Current Assignee: Google Technology Holdings LLC (Alphabet Inc.)
• Original Assignee: Motorola, Inc. (Motorola Solutions, Inc.)
• Inventors: GROBLE, Michael E.
• Granted Patent: US 7,818,271 B2
• Time in Patent Office: Days
• US Class Current: 706/45
• CPC Class Codes: G06N 7/01 Probabilistic graphical models, e.g. probabilistic networks

Abstract

A method and apparatus are disclosed for selecting interaction policies. Values may be provided for a group of parameters for user models. Interaction policies within a specific tolerance of an optimal interaction policy for the user models may be learned. Up to a predetermined number of the learned interaction policies, within a specific tolerance of an optimal policy for the user models, may be selected for a wireless communication device. The wireless communication device, including the selected interaction policies, may determine whether any of a group of parameters, representing a user preference or contextual information with respect to use of the wireless communication device, is updated. When any of the group of parameters has been updated, the wireless communication device may select one of the selected interaction policies, such that the selected one of the selected interaction policies may determine a better interaction behavior for the wireless communication device.

Claims

1. A machine-implemented method for changing an interaction policy in a processing device, the method comprising:

determining whether values of any of a plurality of parameters are updated, each of the plurality of parameters representing at least one of a user preference or contextual information with respect to use of the processing device;

selecting, based on the values of the plurality of parameters, an interaction policy from a plurality of predefined interaction policies included in the processing device, each of the plurality of predefined interaction policies being associated with a respective dialog behavior for the processing device, the selected interaction policy determining a better dialog behavior for the processing device than a previous interaction policy; and

activating the selected interaction policy.

2. The machine-implemented method of claim 1, wherein:

at least some of the plurality of plurality of parameters represent user preferences, and

the user preferences include at least one of an indication of a user'"'"'s patience, an indication of a user'"'"'s tolerance for a wrong dialog action performed by the processing device, and a probability that a user will give up after a speech misrecognition by the processing device.

3. The machine-implemented method of claim 1, further comprising:

receiving data from a user of the processing device with respect to at least one user preference; and

updating a value of at least one of the plurality of parameters as a result of the receiving of the data from the user of the processing device.

4. The machine-implemented method of claim 1, further comprising:

monitoring a recognition accuracy of the processing device;

determining an occurrence of a change in the recognition accuracy of the processing device; and

automatically updating a value of at least one of the plurality of parameters as a result of the determining of the occurrence of the change in the recognition accuracy of the processing device.

5. The machine-implemented method of claim 1, wherein the selecting of the interaction policy from the plurality of predefined interaction policies included in the communication device further comprises:

determining which one of the plurality of predefined interaction policies is a best interaction policy with respect to the plurality of parameters; and

selecting as the interaction policy the one of the plurality of predefined interaction policies determined to be the best interaction policy with respect to the plurality of parameters.

6. The machine-implemented method of claim 5, wherein the determining of which one of the plurality of predefined interaction policies is a best interaction policy with respect to the plurality of parameters further comprises:

determining the best interaction policy based on the plurality of parameters and respective ones of a second plurality of parameters associated with the plurality of predefined interaction policies.

7. A machine-implemented method for selecting up to a predetermined number of a plurality of interaction policies for use on a wireless communication device, the method comprising:

providing sets of values for a plurality of parameters of a plurality of user models, the plurality of user models corresponding to expected users of the wireless communication device, each of the plurality of parameters representing at least one of a user preference or contextual information with respect to use of the processing device;

learning which ones of the plurality of interaction policies are within a specific tolerance of an optimal interaction policy for the plurality of user models; and

selecting, for the wireless communication device, up to the predetermined number of the learned ones of the plurality of interaction policies that are within the specific tolerance of the optimal interaction policy for the plurality of user models.

8. The machine-implemented method of claim 7, wherein the learning of which of the ones of the plurality of interaction policies are within the specific tolerance of the optimal interaction policy for the plurality of user models further comprises:

calculating an average score of each of the plurality of interaction policies with respect to each of the plurality of user models; and

determining which ones of the plurality of interaction policies have a respective average score within the specific tolerance of an optimal interaction policy score with respect to each of the plurality of user models.

9. The machine-implemented method of claim 8, further comprising:

simulating, at least a second predetermined number of times, user actions for each of the plurality of user models and each of the plurality of interaction policies, wherein

the calculating of the score of each of the plurality of interaction policies with respect to each of the plurality of user models is based on the simulating.

10. The machine-implemented method of claim 8, further comprising:

selecting, for the wireless communication device, up to the predetermined number of the learned ones of the plurality of interaction policies that are within the specific tolerance of the optimal interaction policy for the plurality of user models and that are expected to reach a goal of the user in a minimum number of dialog turns.

11. The machine-implemented method of claim 7, wherein the sets of values of the plurality of parameters for the plurality of the plurality of user models are based, at least partly, on market studies.

12. The machine-implemented method of claim 7, wherein:

the plurality of user models is implemented via an interaction model, and

each of the set of values for the plurality of parameters define a user model of the plurality of user models when used with the interaction model.

13. A wireless communication device comprising:

a plurality of predefined interaction policies governing behavior of the wireless communication device when interacting with a user, each of the plurality of predefined interaction policies being associated with values of a parameter set associated with at least one user model;

a parameter updater to update values of a parameter set associated with a current user model; and

a policy selector to select and activate an interaction policy from the plurality of predefined interaction policies based on the updated values of the parameter set associated with the current user model.

14. The wireless communication device of claim 13, wherein:

the policy selector is further arranged to select the interaction policy from the plurality of predefined interaction policies by determining which one of the plurality of predefined interaction policies is a best interaction policy with respect to the parameter set associated with the current user model, and selecting as the interaction policy the one of the plurality of predefined interaction policies determined to be the best interaction policy with respect to the parameter set associated with the current user model.

15. The wireless communication device of claim 14, wherein the policy selector is further arranged to determine which one of the plurality of predefined interaction policies is a best interaction policy with respect to the parameter set associated with the current user model by determining the best interaction policy based on the parameter set associated with the current user model and respective ones of a second plurality of parameters associated with the plurality of predefined interaction policies.

16. The wireless communication device of claim 13, wherein:

the policy selector includes a classifier to classify the parameter set associated with the current user model as being associated with an interaction policy of the plurality of predefined interaction policies based on a second plurality of parameter sets associated with the plurality of predefined interaction policies.

17. The wireless communication device of claim 13, further comprising:

a monitor to monitor use of the wireless communication device, wherein

the parameter updater is arranged to update the values of the parameter set associated with the current user model based on results of monitoring use of the wireless communication device.

18. The wireless communication device of claim 17, wherein:

the monitor is arranged to monitor at least one environmental condition with respect to use of the wireless communication device, and

the parameter updater is arranged to update at least one of the values of the parameter set associated with the current user model when a change in the at least one environmental condition is detected.

19. The wireless communication device of claim 13, wherein:

the parameter updater is arranged to update at least one of the values of the parameter set associated with the current user model in response to received user input.

20. The wireless communication device of claim 19, wherein the received user input includes information with respect to at least one user preference.