Generating and updating conversational artifacts from APIS

Patent metadata

Field	Value
Publication number	US-12124811-B2
Application number	US-202117454302-A
Country	US
Kind code	B2
Filing date	Nov 10, 2021
Priority date	Nov 10, 2021
Publication date	Oct 22, 2024
Grant date	Oct 22, 2024

Abstract

A method, computer system, and a computer program product for generating a conversational bot for an application programming interface (API) is provided. The present invention may include parsing an API schema. The present invention may include generating sentences for the conversational bot from the parsed API schema. The present invention may include constructing the conversational bot by training a deep learning model. The present invention may include receiving a natural language expression from a user. The present invention may include determining whether the natural language expression is enough to activate the bot.

First claim

What is claimed is:

1. A method for generating a conversational bot for an application programming interface (API), the method comprising:
parsing an API schema to extract a plurality of artifacts;
training a deep learning model to generate a plurality of sentences comprising a conversational bot using a plurality of variants of intents and parameters comprising the extracted artifacts;
constructing, by the trained deep learning model, the conversational bot by generating the plurality of sentences;
receiving a natural language expression from a user;
determining whether the natural language expression is enough to activate the bot; and
responsive to determining that the natural language expression is not enough to activate the bot:
retrieving one or more successful utterances from a same session as the natural language expression;
determining a similarity of one or more of the retrieved successful utterances to the natural language expression;
responsive to the similarity of one or more of the retrieved successful utterances exceeding a threshold level, adding the one or more natural language expressions as a sample intent; and
retraining the deep learning model with the sample intent.

2. The method of claim 1 , wherein parsing the API schema further comprises:
identifying a set of descriptions and a summary of what the API does.

3. The method of claim 1 , wherein the sentences are generated for the conversational bot from the parsed API schema using an ensemble of methods.

4. The method of claim 1 , wherein the natural language expression received from the user is a text expression, an utterance or an audible speech.

5. The method of claim 1 , wherein it is determined that the natural language expression is not enough to activate the bot, further comprising:
receiving feedback from the user;
determining a similarity between the received feedback and at least one successful utterance;
associating the natural language expression with the at least one successful utterance in a database; and
retraining the deep learning model.

6. The method of claim 1 , wherein the natural language expression is received in a conversational interface on top of the API.

7. The method of claim 1 , wherein it is determined that the natural language expression is enough to activate the bot, further comprising:
executing a computer code to invoke an API call.

8. A computer system for generating a conversational bot for an application programming interface (API), comprising:
one or more processors, one or more computer-readable memories, one or more computer-readable tangible storage medium, and program instructions stored on at least one of the one or more tangible storage medium for execution by at least one of the one or more processors via at least one of the one or more memories, wherein the computer system is capable of performing a method comprising:
parsing an API schema to extract a plurality of artifacts;
training a deep learning model to generate a plurality of sentences comprising a conversational bot using a plurality of variants of intents and parameters comprising the extracted artifacts;
constructing, by the trained deep learning model, the conversational bot by generating the plurality of sentences;
receiving a natural language expression from a user;
determining whether the natural language expression is enough to activate the bot; and
responsive to determining that the natural language expression is not enough to activate the bot:
retrieving one or more successful utterances from a same session as the natural language expression;
determining a similarity of one or more of the retrieved successful utterances to the natural language expression;
responsive to the similarity of one or more of the retrieved successful utterances exceeding a threshold level, adding the one or more natural language expressions as a sample intent; and
retraining the deep learning model with the sample intent.

9. The computer system of claim 8 , wherein parsing the API schema further comprises:
identifying a set of descriptions and a summary of what the API does.

10. The computer system of claim 8 , wherein the sentences are generated for the conversational bot from the parsed API schema using an ensemble of methods.

11. The computer system of claim 8 , wherein the natural language expression received from the user is a text expression, an utterance or an audible speech.

12. The computer system of claim 8 , wherein it is determined that the natural language expression is not enough to activate the bot, further comprising:
receiving feedback from the user;
determining a similarity between the received feedback and at least one successful utterance;
associating the natural language expression with the at least one successful utterance in a database; and
retraining the deep learning model.

13. The computer system of claim 8 , wherein the natural language expression is received in a conversational interface on top of the API.

14. The computer system of claim 8 , wherein it is determined that the natural language expression is enough to activate the bot, further comprising:
executing a computer code to invoke an API call.

15. A computer program product for generating a conversational bot for an application programming interface (API), comprising:
one or more non-transitory computer-readable storage media and program instructions stored on at least one of the one or more tangible storage media, the program instructions executable by a processor to cause the processor to perform a method comprising:
parsing an API schema to extract a plurality of artifacts;

training a deep learning model to generate a plurality of sentences comprising a conversational bot using a plurality of variants of intents and parameters comprising the extracted artifacts;
constructing, by the trained deep learning model, the conversational bot by generating the plurality of sentences;
receiving a natural language expression from a user;
determining whether the natural language expression is enough to activate the bot; and
responsive to determining that the natural language expression is not enough to activate the bot:
retrieving one or more successful utterances from a same session as the natural language expression;
determining a similarity of one or more of the retrieved successful utterances to the natural language expression;
responsive to the similarity of one or more of the retrieved successful utterances exceeding a threshold level, adding the one or more natural language expressions as a sample intent; and
retraining the deep learning model with the sample intent.

16. The computer program product of claim 15 , wherein parsing the API schema further comprises:
identifying a set of descriptions and a summary of what the API does.

17. The computer program product of claim 15 , wherein the sentences are generated for the conversational bot from the parsed API schema using an ensemble of methods.

18. The computer program product of claim 15 , wherein the natural language expression received from the user is a text expression, an utterance or an audible speech.

19. The computer program product of claim 15 , wherein it is determined that the natural language expression is not enough to activate the bot, further comprising:
receiving feedback from the user;
determining a similarity between the received feedback and at least one successful utterance;
associating the natural language expression with the at least one successf

Assignees

• IBM

Inventors

• Carbajales Sebastian
• Rizk Yara
• Muthusamy Vinod
• Isahagian Vatche
• Mukherjee Kushal
• Huo Siyu
• Reddy Prabhat Maddikunta
• Silva Moran Dario Andres
• Chan Allen Vi Cuong

Classifications

• G06N3/02

  Neural networks · CPC title

• G06N3/08

  Learning methods · CPC title

• G06F18/2178

  based on feedback of a supervisor · CPC title

• G06F18/214

  Generating training patterns; Bootstrap methods, e.g. bagging or boosting · CPC title

• G06F40/205

  Parsing · CPC title