Error context for bot optimization
In one embodiment, an illustrative method herein may comprise: obtaining, by a device, a plurality of indications of errors experienced by a bot performing tasks, wherein each of the plurality of indications includes contextual information of a corresponding error; determining, by the device, correlated errors among the errors experienced by the bot; aggregating, by the device, contextual information of each of the correlated errors into aggregated contextual data; and providing, by the device, the aggregated contextual data with an error notification for a particular correlated error.
1. A method, comprising:
obtaining, by a device, a plurality of indications of errors experienced by a bot performing tasks, wherein each of the plurality of indications includes contextual information and an audio sample of a user request that the bot failed to recognize of a corresponding error;
determining, by the device, correlated errors among the errors experienced by the bot;
aggregating, by the device, contextual information of each of the correlated errors into aggregated contextual data; and
providing, by the device, the aggregated contextual data with an error notification for a particular correlated error.
2. The method as in claim 1 , wherein the contextual information of the corresponding error includes an indication of a product family associated with the corresponding error.
3. The method as in claim 1 , wherein the contextual information of the corresponding error includes an indication of a geographical location associated with the corresponding error.
4. The method as in claim 1 , wherein the contextual information of the corresponding error includes an indication of a team associated with the corresponding error.
5. The method as in claim 1 , wherein the audio sample includes sensitive user data.
6. The method as in claim 5 , wherein the audio sample is obtained from a storage location local to a device of a user in response to an indication of a consent to share from the user.
7. The method as in claim 5 , further comprising:
causing, by the device, a model used by the bot to recognize user intent to be trained using the audio sample.
8. The method as in claim 1 , wherein the aggregated contextual data for the particular correlated error includes an indication of a portion of a total amount of errors related to a particular product family that are of a same type as the particular correlated error.
9. The method as in claim 1 , wherein the aggregated contextual data for the particular correlated error includes an indication of a portion of a total amount of errors originating in a particular geographic location that are of a same type as the particular correlated error.
10. The method as in claim 1 , wherein the aggregated contextual data for the particular correlated error includes an indication of a portion of a total amount of errors related to a particular customer segment that are of a same type as the particular correlated error.
11. The method as in claim 1 , wherein the aggregated contextual data for the particular correlated error includes an indication of a portion of a total amount of errors handled by a particular team that are of a same type as the particular correlated error.
12. The method as in claim 1 , further comprising:
obtaining, by the device, a set of audio samples collected from different users requesting the bot to perform a task, wherein each audio sample of the set of audio samples includes an indication of a user attribute of a user who that audio sample is from; and
identifying, by the device and based on correlations between user attributes of a portion of the different users whose request achieved a particular outcome, a user attribute associated with the particular outcome.
13. The method as in claim 12 , wherein the indication of the user attribute is an indication of a geographic location of the user.
14. The method as in claim 12 , wherein the indication of the user attribute is an indication of a native language of the user.
15. The method as in claim 12 , wherein the indication of the user attribute is an indication of a type of slang language used by the user.
16. The method as in claim 12 , wherein the indication of the user attribute is an indication of an age of the user.
17. A tangible, non-transitory, computer-readable medium having computer-executable instructions stored thereon that, when executed by a processor on a computer, cause the computer to perform a method comprising:
obtaining a plurality of indications of errors experienced by a bot performing tasks, wherein each of the plurality of indications includes contextual information and an audio sample of a user request that the bot failed to recognize of a corresponding error;
determining correlated errors among the errors experienced by the bot;
aggregating contextual information of each of the correlated errors into aggregated contextual data; and
providing the aggregated contextual data with an error notification for a particular correlated error.
18. The tangible, non-transitory, computer-readable medium as in claim 17 , wherein the method further comprises:
identifying, based on the aggregated contextual data for the particular correlated error, a criticality of the particular correlated error.
19. The tangible, non-transitory, computer-readable medium as in claim 17 , wherein the method further comprises:
identifying, based on the aggregated contextual data for the particular correlated error, a remediation technique to prevent future instances of errors of a same type as the particular correlated error.
20. An apparatus, comprising:
one or more network interfaces to communicate with a network;
a processor coupled to the one or more network interfaces and configured to execute one or more processes; and
a memory configured to store a process that is executable by the processor, the process, when executed, configured to:
obtain a plurality of indications of errors experienced by a bot performing tasks, wherein each of the plurality of indications includes contextual information and an audio sample of a user request that the bot failed to recognize of a corresponding error;
determine correlated errors among the errors experienced by the bot;
aggregate contextual information of each of the correlated errors into aggregated contextual data; and
provide the aggregated contextual data with an error notification for a particular correlated error.