Scenario-Intent API Orchestration for Richer Human-Computer Responses

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Current human-machine interaction methods provide simplistic responses that fail to meet user requirements, resulting in a poor dialog effect.

Innovation Solution

A method that involves obtaining multiple application programming interfaces (APIs) based on user scenario intent, user profile, and environment information to provide richer and more comprehensive responses, adapting to different user preferences and environmental conditions.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Loss of information

If multiple APIs are obtained and called based on scenario intent, then the response information becomes richer and more comprehensive, but the system complexity increases

Engineering Contradiction:
Improveresponse information completenessVSAvoidsystem complexity
Core Design Contradiction:
Loss of informationVSDevice complexity

Solution Approach 1:

The patent segments the API selection process into distinct modules: scenario intent recognition module that analyzes user input to determine the intent type, API selection module that maps scenarios to specific APIs, and API execution module that calls the selected APIs. This segmentation allows the system to handle multiple APIs systematically without overwhelming complexity.

Inventive Principle:
Principle #1Segmentation

Solution Approach 2:

The patent introduces an intermediary layer (the scenario intent recognition and API selection mechanism) between user input and API execution. This intermediary analyzes the scenario intent and selectively determines which APIs to call, enabling comprehensive information retrieval while maintaining manageable system complexity through structured mediation.

Inventive Principle:
Principle #24Intermediary (Mediator)

2Adaptability or versatility

If APIs are selected based on user profile and environment information, then the response becomes more personalized and pertinent, but the processing time increases

Engineering Contradiction:
Improvepersonalization capabilityVSAvoidprocessing time
Core Design Contradiction:
Adaptability or versatilityVSLoss of time

Solution Approach 1:

The patent implements preliminary action by pre-establishing the scenario intent recognition framework and API mapping relationships before actual user interactions. The system pre-processes and categorizes different scenario types and their corresponding APIs, so that during actual operation, the system can quickly match user input to pre-defined scenarios and execute the appropriate APIs without extensive real-time analysis.

Inventive Principle:
Principle #10Preliminary action

3Ease of operation

If direct API results are output to the user, then the response convenience is improved, but the response may be too simplistic to meet user requirements

Engineering Contradiction:
Improveresponse convenienceVSAvoidresponse comprehensiveness
Core Design Contradiction:
Ease of operationVSLoss of information

Solution Approach 1:

The patent merges multiple API results into a unified comprehensive response. Instead of outputting results from a single API, the system calls multiple relevant APIs based on the scenario intent and integrates their results into a cohesive response that is both convenient for the user and comprehensive in meeting their requirements.

Inventive Principle:
Principle #5Merging (Combining)

Solution Approach 2:

The patent creates a universal response framework that can handle multiple types of user requirements through a single integrated system. The scenario intent recognition mechanism and multi-API calling capability enable the system to provide comprehensive responses for various scenario types (information queries, operational tasks, personalized services) while maintaining a consistent user-friendly interface.

Inventive Principle:
Principle #6Universality (Multi-functionality)

Data Source

PatentEP4697200A1Human-computer interaction method and electronic device
Publication Date: 2026.02.18 HUAWEI TECH CO LTD
  • EP4697200A1 patent drawingFigure 1(1)
  • EP4697200A1 patent drawingFigure 1(2)
  • EP4697200A1 patent drawingFigure 2~3

AI summary

A human-machine interaction method and an electronic device are provided, and relate to the field of Al. The method is applied to the electronic device, and includes: receiving a user instruction, and when the user instruction indicates a scenario intent, obtaining a plurality of target APIs based on the scenario intent, where the plurality of target APIs correspond to the scenario intent; and then, outputting results of calling the plurality of target APIs, where the results of calling the plurality of APIs may be used to implement a user requirement corresponding to the scenario intent. In this way, the electronic device can directly output, based on a scenario intent of a user, a result that is of calling an API and that meets a user requirement under the scenario intent. The API is a direct exposure of an application function. Results of calling a plurality of APIs are directly output, so that a richer, more comprehensive, and more convenient response result is implemented, to meet a user requirement, and improve a human-machine dialog effect.