Retrieval-Augmented Query Processing for Accurate Health Authority Responses
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional query processing technologies for health authorities are inefficient, relying heavily on human users and subject matter experts, leading to delays in processing thousands of queries and hindering drug development and treatment administration.
Innovation Solution
Employing retrieval augmented generation techniques with a generative ML model to automatically identify relevant documents and generate responses by encoding query text into numeric representations, using similarity measures to match document portions, and prompting the generative ML model with contextual data for accurate responses.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If conventional query processing methods are used, then human experts can ensure accurate responses, but processing speed and efficiency deteriorate due to manual handling of thousands of queries
Solution Approach 1:
The patent introduces an intermediary system comprising a language model and document retrieval mechanism that acts as a mediator between the query and human experts. The system automatically retrieves relevant documents and generates draft responses, which human experts then review and refine. This intermediary processing layer handles the bulk of routine query processing, significantly improving throughput while maintaining response quality through expert oversight.
Solution Approach 2:
The query processing workflow is segmented into distinct automated and manual stages. The system automatically performs document retrieval, text extraction, and draft response generation using machine learning models. Human experts are then assigned only to review and finalize responses requiring judgment. This segmentation allows high-volume routine processing to be automated while preserving human expertise for complex cases.
2Reliability
If more human experts are allocated to process queries, then response quality is maintained, but operational costs and processing time increase
Solution Approach 1:
The system enables self-service query processing through automated document retrieval and response generation capabilities. The language model independently searches databases, extracts relevant information, and drafts responses without requiring human intervention for each query. This self-service automation handles routine processing, reducing both time and cost while human experts focus only on complex cases requiring their specialized knowledge.
3Productivity
If automated processing is implemented, then processing efficiency improves, but response accuracy may deteriorate due to lack of human judgment
Solution Approach 1:
The system dynamically adjusts the level of automation based on query complexity and confidence metrics. For routine queries with clear answers in the database, the system applies full automation. For complex or ambiguous queries, the system automatically identifies these cases and routes them to human experts for handling. This dynamic approach optimizes the balance between automation efficiency and human judgment accuracy.
Data Source
AI summary
Some embodiments relate to a system for processing queries. The system identifies, from among document portions stored in at least one database, at least one document portion to use for responding to request(s) in a query. The system generates, using a generative machine learning (ML) model and the identified document portion(s), a response to the request(s) at least in part by: generating a prompt for the generative ML model using the identified document portion(s); and providing the prompt to the generative ML model to generate the response to the request(s).