System and method of evaluating user simulations in a spoken dialog system with a diversion metric
Systems, methods and computer-readable media associated with using a divergence metric to evaluate user simulations in a spoken dialog system. The method employs user simulations of a spoken dialog system and includes aggregating a first set of one or more scores from a real user dialog, aggregating a second set of one or more scores from a simulated user dialog associated with a user model, determining a similarity of distributions associated with each of the first set and the second set, wherein the similarity is determined using a divergence metric that does not require any assumptions regarding a shape of the distributions. It is preferable to use a Cramér-von Mises divergence.
1. A method comprising:
aggregating, via a processor, a first set comprising a score from a real user dialog and aggregating a second set comprising a score from a simulated user dialog associated with a user model; and
determining a similarity of distributions associated with each of the first set and the second set, wherein the similarity is determined using a divergence metric that does not require any assumptions regarding a shape of the distributions, wherein the divergence metric is normalized.
2. The method of claim 1 , wherein the divergence metric is a Cramér-von Mises divergence.
3. The method of claim 1 , wherein the determined similarity measures a parameter associated with the user model.
4. The method of claim 3 , wherein the parameter measures how good the user model is at simulating real users.
5. The method of claim 3 , further comprising:
modifying one of the user model and the spoken dialog system based on the parameter associated with the user model.
6. A system comprising:
a processor; and
a computer-readable storage medium having instructions stored which, when executed by a computing device, cause the computing device to perform operations comprising:
aggregating a first set comprising a score from a real user dialog and aggregate a second set comprising a score from a simulated user dialog associated with a user model; and
determining a similarity of distributions associated with each of the first set and the second set, wherein the similarity is determined using a divergence metric that does not require any assumptions regarding a shape of the distributions, wherein the divergence metric is normalized.
7. The system of claim 6 , wherein the divergence metric is a Cramér-von Mises divergence.
8. The system of claim 6 , wherein the determined similarity measures a parameter associated with the user model.
9. The system of claim 8 , wherein the parameter measures how good the user model is at simulating real users.
10. The system of claim 8 , the computer-readable storage medium having additional instructions stored which result in the operations further comprising modifying one of the user model and the spoken dialog system based on the parameter associated with the user model.
11. A computer-readable storage device having instructions stored which, when executed by a computing device, cause the computing device to perform operations comprising:
aggregating a first set comprising a score from a real user dialog and aggregating a second set comprising a score from a simulated user dialog associated with a user model; and
determining a similarity of distributions associated with each of the first set and the second set, wherein the similarity is determined using a divergence metric that does not require any assumptions regarding a shape of the distributions, wherein the divergence metric is normalized.
12. The computer-readable storage device of claim 11 , wherein the divergence metric is a Cramér-von Mises divergence.
13. The computer-readable storage device of claim 11 , wherein the determined similarity measures a parameter associated with the user model.
14. The computer-readable storage device of claim 11 , the computer-readable storage device having additional instructions stored which result in the operations further comprising:
modifying one of the user model and the spoken dialog system based on the parameter associated with the user model.