Network Query Time Disambiguation Using Schema Metadata
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Natural language queries in network monitoring systems often lead to ambiguities regarding time windows and data aggregation, resulting in irrelevant or incorrect answers due to the complexity of network telemetry data stored as time series.
Innovation Solution
A device retrieves schema metadata from a database to resolve time ambiguities in queries using a large language model, allowing for precise time disambiguation and generating accurate answers.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Ease of operation
If natural language queries are used for network monitoring, then ease of operation is improved, but measurement precision deteriorates due to time ambiguities
Solution Approach 1:
The patent introduces an intermediary component (time disambiguation module or LLM) that sits between the user's natural language query and the network telemetry data retrieval system. This intermediary automatically resolves time ambiguities by inferring the intended time window from context, schema metadata, or user preferences, thereby maintaining ease of operation while ensuring measurement precision without requiring users to manually specify time parameters
Solution Approach 2:
The system performs preliminary action by pre-defining time windows and aggregation parameters based on schema metadata and contextual information before the actual query execution. This allows the system to automatically resolve time ambiguities in advance, so that when natural language queries are processed, the time parameters are already determined, ensuring both ease of operation and measurement precision
2Reliability
If time series data is stored for network telemetry, then reliability is improved, but device complexity increases
Solution Approach 1:
The patent applies preliminary action by pre-organizing time series data with associated schema metadata that defines time windows, aggregation methods, and data structures. This preliminary organization enables reliable time-based queries without requiring complex real-time processing, as the data is already structured for efficient retrieval and analysis
Solution Approach 2:
The system introduces an intermediary layer (time disambiguation module) that manages the complexity of time series data retrieval. This intermediary handles the complex logic of time window determination and data aggregation, shielding users from complexity while ensuring reliable access to historical network telemetry data
3Productivity
If natural language queries are processed without time disambiguation, then productivity is improved, but loss of information increases due to ambiguous time windows
Solution Approach 1:
The patent implements feedback mechanisms where the system analyzes the natural language query, identifies time ambiguities, and either automatically resolves them using contextual feedback from schema metadata and user preferences, or presents disambiguation options to the user. This feedback loop ensures that no information is lost due to ambiguous time windows while maintaining high productivity through automated resolution
Solution Approach 2:
An intermediary time disambiguation module processes natural language queries to extract and clarify time window specifications. This intermediary prevents information loss by ensuring that time parameters are properly determined before data retrieval, while maintaining productivity by automating the disambiguation process rather than requiring manual clarification
Data Source
AI summary
A device may obtain a query from a user interface regarding a computer network. The device may retrieve, based on the query, schema metadata from a database. The device may resolve any time ambiguity in the query using the schema metadata. The device may provide an answer to the query to the user interface, after resolving any time ambiguity in the query using the schema metadata.