Federated Query Generation Across Product Data Silos
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing techniques for generating queries are limited in their ability to access and retrieve data from multiple data silos, often requiring manual intervention and failing to provide a flexible, automated solution for diverse data sources.
Innovation Solution
A method for fine-tuning a pre-trained language model using domain-specific ontology and vocabulary metadata to generate federated queries, enabling automated access to multiple data silos through natural language prompts.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If manual query generation techniques are used for accessing data silos, then query accuracy can be maintained, but productivity and ease of operation deteriorate due to manual intervention requirements
Solution Approach 1:
The system enables self-service by allowing users to directly query multiple data silos through natural language without requiring manual query construction or expert intervention. The federated query system automatically handles data retrieval across distributed sources based on user-friendly natural language inputs.
Solution Approach 2:
The patent replaces manual mechanical query construction processes with automated language model-based generation. Instead of requiring users to manually construct complex queries, the system uses AI models to automatically generate and execute queries across data silos based on natural language prompts.
2Ease of operation
If centralized data storage is implemented, then ease of operation improves through unified access, but loss of information increases due to data silos being forced into single location
Solution Approach 1:
The system introduces a federated query system as an intermediary layer that enables unified access to data across multiple silos without consolidating the data itself. This mediator allows queries to be routed to appropriate data sources while maintaining data independence and preventing information loss.
Solution Approach 2:
The federated query system provides universal access capabilities that work across multiple different data silos and sources. Instead of requiring separate access mechanisms for each data silo, the system offers a unified interface that can query any connected data source while preserving their original integrity.
3Adaptability or versatility
If existing query generation models are used, then ease of manufacture is maintained, but adaptability deteriorates when accessing diverse data silos with different structures
Solution Approach 1:
The system adapts to different data silos by dynamically adjusting query parameters and structures based on the specific data source being accessed. The federated query system modifies its behavior and query formulation according to the schema and structure of each connected data silo, enabling versatility without requiring complete model redesign.
Data Source
AI summary
Various examples of the disclosure generally relate to federated queries for accessing data from multiple data silos that are associated with a product, such as a medical imaging device. Various examples of the disclosure more specifically relate to a language model for generating such federated queries. Various examples of the disclosure also more specifically relate to fine-tuning such language model.