Natural language-aided conversational exploration of self-documenting application programming interfaces

Patent metadata

Field	Value
Publication number	US-11036728-B2
Application number	US-201916532966-A
Country	US
Kind code	B2
Filing date	Aug 6, 2019
Priority date	Jul 17, 2019
Publication date	Jun 15, 2021
Grant date	Jun 15, 2021

Abstract

Intent-based exploring of self-documenting Application Programming Interface (“API”) documents comprises receiving a document containing a plurality of descriptions of APIs, wherein the document is in a structured format; generating a plurality of entities by identifying a plurality of API definition entries in the document; generating a plurality of intents by iterating through a plurality of path entries in the document; generating a dialog by creating a plurality of phrases by iterating through the plurality of path entries in the document; receiving one or more questions from a user or developer; and in response to the one or more questions, providing one or more conversational responses to the user by a conversational application program incorporating the plurality of entities, plurality of intents, and plurality of phrases.

First claim

What is claimed is:

1. A method for intent-based exploring a self-documenting Application Programming Interface (“API”) document, the method comprising:
receiving, by a computer, a document containing a plurality of descriptions of APIs, wherein the document is in a structured format;
generating, by the computer, a plurality of entities, wherein each entity of the plurality of entities is associated with an API definition entry of the plurality of API definition entries in the document and wherein each entity of the plurality of entities is generated by replacing the properties attribute of the API definition entry with a regex wildcard;
generating, by the computer, a plurality of intents by iterating through the plurality of entities;
generating, by the computer, a dialog by creating a plurality of phrases by iterating through the plurality of path entities in the document;
receiving, by the computer, one or more questions from a user; and
in response to the one or more questions, providing one or more conversational responses to the user by a conversational application program incorporating the plurality of entities, plurality of intents, and plurality of phrases;
thereby improving efficiency of a user interface for a software developer tool to find suitable APIs, and improving efficiency of a runtime computing environment by improving the selection of APIs which consume minimal appropriate computing and communications resources.

2. The method as set forth in claim 1 wherein the generating of a plurality of intents by iterating through a plurality of path entries in the document comprises, for each path:
retrieving operation ID values;
retrieving operation details to generate intent examples; and
creating an intent from the operation ID values and the generated intent examples.

3. The method as set forth in claim 1 wherein the generating of a dialog by creating a plurality of phrases by iterating through the plurality of path entries in the document comprises, for each path:
creating exploring API dialog nodes, and
creating calling API dialog nodes.

4. The method as set forth in claim 1 wherein the self-documenting Application Programming Interface (“API”) document comprises a document selected from the group consisting of a Swagger-compliant document and an OpenAPI-compliant document.

5. The method as set forth in claim 1 wherein the self-documenting Application Programming Interface (“API”) document comprises a REpresentational State Transfer (RESTful) application programming interface description document.

6. The method as set forth in claim 1 , wherein the regex wildcard is a synonym.

Assignees

• IBM

Inventors

• Torres Isa M
• Freed Andrew R

Classifications

• G06F16/243 Primary

  Natural language query formulation · CPC title

• G06F16/24522

  Translation of natural language queries to structured queries · CPC title

• G06F40/40

  Processing or translation of natural language (natural language analysis G06F40/20; semantic analysis G06F40/30) · CPC title

• G06F16/24578

  using ranking · CPC title

• G06F40/30 Primary

  Semantic analysis · CPC title