Back-end task fulfillment for dialog-driven applications
A determination is made as to whether a value of a first parameter of a first application is to be obtained using a natural language interaction. Based on received input, a first service of a plurality of services is identified. The first service is to be used to perform a first task associated with the first parameter. Portions of the first application to determine the value of the first parameter and to invoke the first service are generated.
1. A computer-implemented method, comprising:
prior to depolying an application to make it available for a user:
obtaining, via one or more programmatic interfaces of a dialog-driven-application management service of a cloud computing environment, a request to define a first intent of the application, wherein the first intent corresponds to a first task to be initiated on behalf of the user in response to one or more dialog inputs obtained from the user via on or more end user-related programmatic interfaces;
storing an indication of one or more parameters of the first intent; wherein the first task is to be performed based on the one or more parameters of the first internet;
provisioning, in the cloud computing environment, one or more computer resources to perform a least a first portion of the first task wherein the one or more compute resources includes one or more virtual machines allocated for the application, and
generating integration code to integrate the application with a fulfillment resource external to the cloud computing environment, wherein the fulfillment resource is configured to perform a least a second portion of the first task; and
after the application is deployed:
receiving, from the application, one or more respective values of the one or more parameters of the first intent determined based at least in part on processing of a dialog input from the user using one or more machine learning models, wherein the one or more machine learning models generates a match score of the dialog input with respect to the first intent and the match score exceeds a match threshold,
causing the one or more compute resources of the cloud computing environment to perform the first portion of the first task based on the respective values of the one or more parameters; and
causing the second portion of the first task to be performed at by the fulfillment resource external to the cloud computing environment.
2. The computer-implemented method as recited in claim 1 , further comprising:
obtaining, via the one or more programmatic interfaces, an indication of another service of the cloud computing environment which is to be used to perform the first portion of the first task, and
causing, during the execution of the application, the first portion of the first task to be performed at the other service.
3. The computer-implemented method as recited in claim 1 , further comprising:
obtaining, via the one or more programmatic interfaces, an indication of a sequence in which a plurality of resources are to be employed to perform one or more tasks of the application; and
causing, during the execution of the application, the one or more tasks to be performed at the plurality of resources in the indicated sequence.
4. The computer-implemented method as recited in claim 1 , further comprising:
obtaining, via the one or more programmatic interfaces, an indication of one or more initiation words for a dialog of the application, and
initiating execution of the application based at least in part on a detection, within input provided by the user, of an initiation word of the one or more initiation words.
5. The computer-implemented method as recited in claim 1 , wherein the integration code transmits one or more commands to a web service interface.
6. The computer-implemented method as recited in claim 1 , further comprising:
obtaining, via the one or more programmatic interfaces, an indication of a number of attempts to be made to determine a value of a first parameter of the one or more parameters before abandoning a dialog interaction of the application.
7. The computer-implemented method as recited in claim 1 , further comprising:
inserting, into a lexicon associated with the application, one or more words learned by a natural language understanding model from log records of user interactions with the application, and
utilizing the one or more learned words in one or more other user interactions.
8. A system, comprising:
one or more computing devices,
wherein the one or more computing devices include instructions that upon execution on or across the one or more computing devices cause the one or more computing devices to:
prior to deployment of an application to make it available for a user:
obtain, via one or more development-related programmatic interfaces of a dialog-driven application management service of a cloud computing environment, a request to define a first intent of the application, wherein the first intent corresponds to a first task to be initiated on behalf of the user in response to one or more dialog inputs obtained from the user via one or more end user-related programmatic interfaces,
store an indication of one or more parameters of the first intent, wherein the first task is to be performed based on the one or more parameters of the first intent;
provision, in the cloud computing environment, one or more compute resources to perform a least a first portion of the first task, wherein the one or more compute resources includes one or more virtual machines allocated for the application, and
generating integration code to integrate the application with a fulfillment resource external to the cloud computing environment, wherein the fulfillment resource is configured to perform a least a second portion of the first task; and
after the deployment of the application:
receive, from the application, one or more respective values of the one or more parameters of the first intent determined based at least in part on processing of a dialog input from the user using one or more machine learning model, wherein the one or more machine learning models generates a match score of the dialog input with respect to the first intent and the match score exceeds a match threshold,
cause the one or more compute resources of the cloud computing environment to perform the first portion of the first task based on the respective values of the one or more parameters; and
cause the second portion of the first task to be performed at by the fulfillment resource external to the cloud computing environment.
9. The system as recited in claim 8 , wherein the one or more computing devices include further instructions that upon execution on or across the one or more computing devices further cause the one or more computing devices to:
obtain, via the one or more programmatic interfaces, an indication of another service of the cloud computing environment which is to be used to perform the first portion of the first task, and
initiate, during the execution of the application, the first portion of the first task at the other service.
10. The system as recited in claim 8 , wherein the one or more computing devices include further instructions that upon execution on or across the one or more computing devices further cause the one or more computing devices to:
obtain, via the one or more programmatic interfaces, an indication of a sequence in which a plurality of resources are to be employed to perform one or more tasks of the application, and
initiate, during the execution of the application, the one or more tasks at the plurality of resources in the indicated sequence.
11. The system as recited in claim 8 , wherein the one or more computing devices include further instructions that upon execution on or across the one or more computing devices further cause the one or more computing devices to:
obtain, via the one or more programmatic interfaces, an indication of one or more initiation words for a dialog of the application; and
initiate execution of the application based at least in part on a detection, within input provided by the user, of an initiation word of the one or more initiation words.
12. The system as recited in claim 8 , wherein the integration code transmits one or more commands to a web service interface.
13. The system as recited in claim 8 , wherein the one or more computing devices include further instructions that upon execution on or across the one or more computing devices further cause the one or more computing devices to:
obtain, via the one or more programmatic interfaces, an indication of a number of attempts to be made to determine a value of a first parameter of the one or more parameters before abandoning a dialog interaction of the application.
14. The system as recited in claim 8 , wherein the one or more computing devices include further instructions that upon execution on or across the one or more computing devices further cause the one or more computing devices to:
inserting, into a lexicon associated with the application, one or more words learned by a natural language understanding model from log records of user interactions with the application, and utilizing the one or more learned words in one or more other user interactions.
15. One or more non-transitory computer-accessible storage media storing program instructions that when executed on or across one or more processors cause the one or more processors to:
prior to deployment of an application to make it available for a user:
obtain, via one or more development-related programmatic interfaces of a dialog-driven application management service of a cloud computing environment, a request to define a firstintent of the application, wherein the firstintent corresponds to a firsttask to be initiated on behalf of the user in response to one or more dialog inputs obtained from the user via one or more end user-related programmatic interfaces;
store an indication of one or more parameters of the firstintent, wherein the first task is to be performed based on the one or more parameters of the first intent,
provision, in the cloud computing environment, one or more compute resources to perform a least a first portion of the first task, wherein the one or more compute resources includes one or more virtual machines allocated for the application, and
generate integration code to integrate the application with a fulfillment resource external to the cloud computing environment, wherein the fulfillment resource is configured to perform a least a second portion of the first task; and
after the deployment of the application:
receive, from the application, one or more respective values of the one or more parameters of the firstintent determined based at least in part on processing of a dialog input from the user using one or more machine learning models, wherein the one or more machine learning models generates a match score of the dialog input with respect to the firstintent and the match score exceeds a match threshold;
cause the one or more compute resources of the cloud computing environment to perform the first portion of the first task based on the respective values of the one or more parameters; and
cause the second portion of the firsttask to be performed at by the fulfillment resource external to the cloud computing environment.
16. The one or more non-transitory computer-accessible storage media as recited in claim 15 , storing further program instructions that when executed on or across one or more processors further cause the one or more processors to:
obtain, via the one or more programmatic interfaces, an indication of another service of the cloud computing environment which is to be used to perform the first portion of the first task, and
initiate, during the execution of the application, the first portion of the first task at the other service.
17. The one or more non-transitory computer-accessible storage media as recited in claim 15 , storing further program instructions that when executed on or across one or more processors further cause the one or more processors to:
obtain, via the one or more programmatic interfaces, an indication of a sequence in which a plurality of resources are to be employed to perform one or more tasks of the application; and
initiate, during the execution of the application, the one or more tasks at the plurality of resources in the indicated sequence.
18. The one or more non-transitory computer-accessible storage media as recited in claim 15 , storing further program instructions that when executed on or across one or more processors further cause the one or more processors to:
obtain, via the one or more programmatic interfaces, an indication of one or more initiation words for a dialog of the application, and
initiate execution of the application based at least in part on a detection, within input provided by the user, of an initiation word of the one or more initiation words.
19. The one or more non-transitory computer-accessible storage media as recited in claim 15 , wherein the integration code transmits one or more commands to a web service interface.
20. The one or more non-transitory computer-accessible storage media as recited in claim 15 , storing further program instructions that when executed on or across one or more processors further cause the one or more processors to:
obtain, via the one or more programmatic interfaces, an indication of a number of attempts to be made to determine a value of a firstparameter of the one or more parameters before abandoning a dialog interaction of the application.