Virtual services rapid deployment tool

• US 10,521,462 B2
• Filed: 02/27/2018
• Issued: 12/31/2019
• Est. Priority Date: 02/27/2018
• Status: Active Grant

• Current Assignee: Accenture Global Solutions Limited (Accenture PLC)
• Original Assignee: Accenture Global Solutions Limited (Accenture PLC)
• Inventors: Acampado, Carl Thomas Marquises
• Primary Examiner(s): Gauthier, Gerald
• Application Number:
  US15/907,111
• Publication Number:
  US 20190266280A1
• Time in Patent Office: 672 Days
• Field of Search: 704 9, 704275, 704 7, 704257, 704 8, 705 2, 705 5, 705 261, 707769, 709206, 709223, 714 3814, 717124, 717100, 726 19, 379 3201, 702184, 706 45, 706 11, 706 55, 3405721
• CPC Class Codes: G06F 16/3329 Natural language query formulation or dialogue systems
• CPC Class Codes: G06F 40/00 Handling natural language data speech analysis or synthesis, speech recognition G10L
• CPC Class Codes: G06F 40/35 Discourse or dialogue representation
• CPC Class Codes: G06F 8/35 model driven
• CPC Class Codes: G06F 8/60 Software deployment
• CPC Class Codes: G06F 9/455 Emulation; Interpretation; Software simulation, e.g. virtualisation or emulation of application or operating system execution engines
• CPC Class Codes: H04L 67/10 in which an application is distributed across nodes in the network software deployment G06F8/60; multiprogramming arrangements G06F9/46

Abstract

A system for rapid deployment and configuration of a virtual agent to a hosted cloud platform is provided. A virtual agent builder may convert a dialog corpus into a dialog model that includes parameters that cause a cognitive service engine to provide natural language processing for the virtual agent according to the dialog corpus. Depending on the natural language processing capabilities of the cognitive engine service, the dialog model may be included in the virtual agent or communicated to the cognitive service engine. The virtual agent, or individual components of the virtual agent, may be generated and/or reconfigured in real time using predetermined logic that is selected based on virtual agent configuration parameters. The virtual agent configuration parameters may be received from a graphical user interface that reduces the complexity of deploying and/or managing the virtual agent.

Claims

1. A system, the system comprising:

a memory comprising instructions for deployment of a virtual agent; and

a processor, the processor executing the instructions for deployment of the virtual agent that cause the processor to;

generate a graphical user interface, the graphical user interface comprising a first control, a second control, and a third control, the first control configured to enable selection of a cognitive service engine from a plurality of cognitive service engines, the second control configured to receive a dialog corpus, the third control configured to generate a trigger to deploy the virtual agent;

obtain a service identifier of the selected cognitive service engine indicated by the first control;

obtain the dialog corpus received by the second control, the dialog corpus comprising parameters in a first structured format, the parameters comprising a plurality of intent parameters and a plurality of intent response parameters, and a mapping between at least one of the intent parameters and at least one of the intent response parameters;

generate a virtual agent for a hosted-cloud platform, the virtual agent including instructions to receive natural language input and to respond the natural language input using the selected cognitive service engine;

select, based on the service identifier, a service builder stored in a database, the database including a plurality of service builders configured to generate respective dialog models, the respective dialog models independently configured for separate cognitive service engines, wherein the selected service builder is associated with a service builder logic, the service builder logic comprising an instruction to combine at least one of the parameters of the dialog corpus with template logic associated with the service builder;

generate, by execution of the service builder logic associated with the selected service builder, a dialog model based on combination of the template logic with the at least one of the parameters of the dialog corpus, the dialog modal comprising a structured format different from the first structured format, the second structured format configured for the selected cognitive service engine, the dialog model including parameters which cause the selected cognitive service engine to process natural language input according to the intent parameters and the intent response parameters included in the dialog corpus;

depending on the selected cognitive service engine and the service builder, either configure the selected cognitive service engine with the generated dialog model by communication of the generated dialog model to the cognitive service engine, or generate a dialog model service application that uses the generated dialog model to interact with the selected cognitive service engine; and

deploy, in response to detection of the trigger from the third control, the virtual agent to the hosted-cloud platform.

2. The system of claim 1, wherein the instructions for deployment of the virtual agent further cause the processor to:

generate a diagnostics controller for the virtual agent, the diagnostics controller configured to test the virtual agent based on a test case, wherein the test case communicates a test message to the virtual agent based on at least one of the intent parameters of the dialog corpus, wherein the test case compares an output message of the virtual agent to an expected result based on at least one of the intent response parameters of the dialog corpus.

3. The system of claim 2, wherein the instructions that cause the processor to generate a diagnostics controller for the virtual agent further comprise instructions that cause the processor to:

access test case logic stored in a logic repository, the test case logic comprising an input field and a result field;

replace the input field with intent request information corresponding to at least one of the intent parameters of the dialog corpus;

replace the result field with intent response information corresponding to at least one of the intent response parameters of the dialog corpus; and

include the test case logic configured with the intent request information and intent response information in a diagnostics controller of the virtual agent, wherein the diagnostics controller executes the test case logic to test the virtual agent based on the intent request information and the intent response information.

4. The system of claim 1, wherein the instructions for deployment of the virtual agent further cause the processor to:

receive, from the graphical user interface, an analytics type identifier;

select analytics engine logic stored in a logic repository, the analytics logic associated with the analytics type identifier;

generate a real-time analytics engine for the virtual agent, the real-time analytics engine including the selected analytics engine logic, wherein the analytic engine logic causes the real-time analytics engine to monitor one or more variables and display the one or more variables on the graphical user interface.

5. The system of claim 1, wherein the selected cognitive service engine provides natural language generation, wherein the instructions for deployment of the virtual agent further cause the processor to configure the selected cognitive service engine with the generated dialog model by communication of the generated dialog model to the cognitive service engine.

6. The system of claim 1, wherein the dialog model comprises instructions that cause the virtual agent to:

communicate, to the selected cognitive service engine, natural language input received from a terminal device;

receive, from the cognitive service engine, an intent classification corresponding to the natural language input; and

communicate, to the terminal device that sent the natural language input, a response based on the dialog corpus.

7. A method to manage generation and deployment of a virtual agent to a hosted could platform, the method comprising:

generating a graphical user interface, the graphical user interface comprising a first control, a second control, and a third control, the first control configured to enable selection of a cognitive service engine from a plurality of cognitive service engines, the second control configured to receive a dialog corpus, the third control configured to generate a trigger to deploy the virtual agent;

obtaining a service identifier of the selected cognitive service engine indicated by the first control;

obtaining the dialog corpus received by the second control, the dialog corpus comprising a plurality of intent parameters, a plurality of intent response parameters, and a mapping between at least one of the intent parameters and at least one of the intent response parameters, the intent parameters, the intent response parameters, and the mapping being formatted in a first structured format;

generating the virtual agent for a hosted-cloud platform, the virtual agent including instructions to receive natural language input and to respond the natural language input using the selected cognitive service engine;

selecting, based on the service identifier, a service builder stored in a database, the database including a plurality of service builders configured to generate respective dialog models, the respective dialog models independently configured for separate cognitive service engines, wherein the selected service builder is associated with a service builder logic, the service builder logic comprising an instruction to combine at least one of the parameters of the dialog corpus with template logic associated with the service builder;

generating, by execution of the service builder logic associated with the selected service builder, a dialog model based on combination of the template logic with the at least one of the parameters of the dialog corpus, the dialog modal comprising a structured format different from the first structured format, the second structured format configured for the selected cognitive service engine, the dialog model including parameters which cause the selected cognitive service engine to process natural language input according to the intent parameters and the intent response parameters included in the dialog corpus;

depending on the selected cognitive service engine and the service builder, either

configuring the selected cognitive service engine with the generated dialog model by communication of the generated dialog model to the cognitive service engine, or

generating a dialog model service application that uses the generated dialog model to interact with the selected cognitive service engine; and

deploying, in response to detection of the trigger from the third control, the virtual agent to the hosted-cloud platform.

8. The method of claim 7, further comprising:

generating a diagnostics controller for the virtual agent, the diagnostics controller configured to test the virtual agent based on a test case, wherein the test case communicates a test message to the virtual agent based on at least one of the intent parameters of the dialog corpus, wherein the test case compares an output message of the virtual agent to an expected result based on at least one of the intent response parameters of the dialog corpus.

9. The method of claim 8, wherein generating the diagnostics controller for the virtual agent further comprises:

accessing test case logic stored in a repository, the test case logic comprising an input field and a result field;

replacing the input field with intent request information corresponding to at least one of the intent parameters of the dialog corpus;

replacing the result field with intent response information corresponding to t at least one of the intent response parameters of the dialog corpus; and

including the test case logic configured with the intent request information and intent response information in a dialog controller of the virtual agent, wherein the dialog controller executes the test case logic to test the virtual agent.

10. The method of claim 7, further comprising:

receiving a cognitive service update request comprising a second service identifier corresponding to a second cognitive service engine;

selecting, based on the second service identifier, a second service builder corresponding to the second cognitive service engine;

generating, by execution of service builder logic associated with the second service builder, a second dialog model configured for the second cognitive service engine; and

reconfiguring the dialog model service application to interact with the second cognitive service engine based on the second dialog mode.

11. The method of claim 7, wherein the selected cognitive service engine is configured to provide natural language generation, wherein the method further comprises:

configuring the selected cognitive service engine with the generated dialog model by communication of the generated dialog model to the cognitive service engine.

12. The method of claim 7, wherein the dialog model comprises instructions that cause the virtual agent to:

communicate, to the selected cognitive service engine, natural language input received from a terminal device;

receive, from the cognitive service engine, an intent classification corresponding to the natural language input; and

communicate, to the terminal device that sent the natural language input, a response based on the dialog corpus.

13. Circuitry for rapidly deployment of a virtual agent to a hosted could platform, the circuitry comprising:

circuitry configured to generate a graphical user interface, the graphical user interface comprising a first control, a second control, and a third control, the first control configured to enable selection of a cognitive service engine from a plurality of cognitive service engines, the second control configured to receive a dialog corpus, the third control configured to generate a trigger to deploy the virtual agent;

circuitry configured to obtain a service identifier of the selected cognitive service engine indicated by the first control;

circuitry configured to obtain the dialog corpus from the second control, the dialog corpus comprising a plurality of intent parameters, a plurality of intent response parameters, and a mapping between at least one of the intent parameters and at least one of the intent response parameters, the intent parameters, the intent response parameters, and the mapping being formatted in a first structured format;

circuitry configured to generate the virtual agent for a hosted-cloud platform, the virtual agent including instructions to receive natural language input and to respond the natural language input using the selected cognitive service engine;

circuitry configured to select, based on the service identifier, a service builder stored in a database, the database including a plurality of service builders configured to generate respective dialog models, the respective dialog models independently configured for separate cognitive service engines, wherein the selected service builder is associated with service builder logic, the service builder logic comprising an instruction to combine at least one of the parameters of the dialog corpus with template logic associated with the service builder;

circuitry configured to generate, by execution of the service builder logic associated with the selected service builder, a dialog model based on combination of the template logic with the at least one of the parameters of the dialog corpus, the dialog modal comprising a structured format different from the first structured format, the second structured format configured for the selected cognitive service engine, the dialog model including parameters which cause the selected cognitive service engine to process natural language input according to the intent parameters and the intent response parameters included in the dialog corpus;

circuitry configured to depending on the selected cognitive service engine and the service builder, either configure the selected cognitive service engine with the generated dialog model by communication of the generated dialog model to the cognitive service engine, or generate a dialog model service application that uses the generated dialog model to interact with the selected cognitive service engine; and

circuitry configured to deploy, in response to detection of the trigger from the third control, the virtual agent to the hosted-cloud platform.

14. The circuitry of claim 13, further comprising:

circuitry configured to generate a diagnostics controller for the virtual agent, the diagnostics controller configured to test the virtual agent based on a test case, wherein the test case communicates a test message to the virtual agent based on at least one of the intent parameters of the dialog corpus, wherein the test case compares an output message of the virtual agent to an expected result based on at least one of the intent response parameters of the dialog corpus.

15. The circuitry of claim 14, wherein the circuitry configured to generate the diagnostics controller for the virtual agent further comprises:

circuitry configured to access test case logic stored in a repository, the test case logic comprising a an input field and a result field;

circuitry configured to replace the input field with intent request information corresponding to at least one of the intent parameters of the dialog corpus;

circuitry configured to replace the result field with intent response information corresponding to t at least one of the intent response parameters of the dialog corpus; and

circuitry configured to include the test case logic configured with the intent request information and intent response information in a dialog controller of the virtual agent, wherein the dialog controller executes the test case logic to test the virtual agent.

16. The circuitry of claim 13, further comprising:

circuitry configured to communicate the generated dialog model to the cognitive service engine.

17. The circuitry of claim 13, wherein the dialog model comprises instructions that cause the virtual agent to:

communicate, to the selected cognitive service engine, natural language input received from a terminal device;

receive, from the cognitive service engine, an intent classification corresponding to the natural language input; and

communicate, to the terminal device that sent the natural language input, a response based on the dialog corpus.