Intent Model Interface for Cross-Vendor Network Management
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Complex networks with diverse services and rapid updates in communications technologies lead to increased operational and management difficulties, necessitating complex and costly network management, and hinder cross-vendor integration due to numerous intent management interfaces.
Innovation Solution
Implementing a universal Intent-Driven Management Service (IDMS) interface that configures intent models comprising intent objects, actions, and parameters, allowing flexible and evolvable management across different scenarios.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If multiple processing components (NLP, knowledge graph, reasoning engines) are integrated to improve intent processing accuracy, then the processing accuracy and completeness improve, but the system complexity increases
Solution Approach 1:
The system divides intent processing into distinct modular components: NLP processing unit, knowledge graph processing unit, rule-based reasoning engine, and machine learning model. Each component handles specific aspects of intent analysis independently, improving accuracy through specialized processing while allowing the system to manage complexity through clear separation of functions.
Solution Approach 2:
The patent introduces an intermediary layer that coordinates between multiple processing components. This intermediary manages the flow of data between NLP, knowledge graph, and reasoning engines, integrating their outputs to form comprehensive intent understanding without requiring direct complex interactions between all components.
2Ease of operation
If comprehensive user context and historical data are analyzed to improve service personalization, then the user experience improves, but the processing time increases
Solution Approach 1:
The system performs preliminary processing of user context and historical data by maintaining pre-built knowledge graphs and user profiles that store processed information in advance. When a user query arrives, the system queries these pre-processed structures rather than analyzing raw historical data from scratch, significantly reducing processing time while still providing personalized service.
3Reliability
If multiple data sources and processing methods are integrated to improve processing completeness, then the intent understanding improves, but the computational resources required increase
Solution Approach 1:
The system implements partial processing by selectively applying different processing methods based on the specific intent type and available data. Not all queries require full NLP processing, knowledge graph lookup, and machine learning analysis simultaneously. The system applies only the necessary processing depth and breadth for each specific case, reducing computational resource consumption while maintaining accurate intent understanding for each query type.
Data Source
Figure 1~2
Figure 2a~3
Figure 4A
AI summary
Embodiments of this application provide an intent processing method and apparatus, and a system, to improve architecture flexibility and evolvability. The method includes: A management device receives an intent from a requirement device through an IDMS interface, where the intent includes information about an intent object, information about an intent action, and information about an intent-specific parameter, and the intent is used to indicate the management device to perform, on an object indicated by the information about the intent object, an action indicated by the information about the intent action. The management device obtains a configuration parameter based on the received intent. The management device configures, based on the obtained configuration parameter, a network or a service managed by the management device.