LLM-Generated Summaries for Semantic Structured Data Search
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional keyword-based searches for structured data in cloud platforms often lose context and yield inaccurate results due to the reliance on specific query phrases, making it difficult to efficiently retrieve structured data from large volumes of information.
Innovation Solution
A system utilizing generative AI and a large language model (LLM) converts structured metadata into unstructured summaries, enabling semantic searching by generating natural language summaries that capture the intent of the data, which are then vectorized and compared to user queries for accurate retrieval.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Ease of operation
If keyword-based search is used for structured data, then search simplicity is maintained, but search accuracy and context retention deteriorate
Solution Approach 1:
The patent introduces an intermediary component (semantic search engine) that translates natural language queries into structured data searches. This mediator layer allows users to search using simple natural language while the system handles the complex matching against structured metadata, resolving the contradiction between ease of operation and search accuracy.
Solution Approach 2:
The system changes the parameter of query formulation from requiring precise keywords to accepting natural language. By transforming the search interface to work with unstructured natural language inputs rather than structured keywords, the system improves ease of operation while maintaining accuracy through underlying structured metadata comparison.
2Measurement precision
If specific query phrases are required for semantic search, then search precision is improved, but user accessibility and ease of use deteriorate
Solution Approach 1:
The system enables users to perform semantic searches without requiring them to learn specific query phrases or understand the underlying data structure. The semantic search engine automatically analyzes natural language queries and matches them against structured metadata, allowing users to search effectively without specialized knowledge.
3Device complexity
If keyword-based search is used, then system complexity is reduced, but information loss and context deterioration occur
Solution Approach 1:
The patent applies preliminary action by pre-processing and structuring metadata from unstructured data sources before the search operation. This advance preparation of structured metadata enables accurate semantic searching without requiring complex real-time processing during the actual search, maintaining system simplicity while preserving context.
Data Source
AI summary
Methods, systems, apparatuses, devices, and computer program products are described. An application server or a data processing system may convert a set of metadata associated with a data object (e.g., document, record, asset) from a first structured format into a second serialized format. The set of metadata in the second serialized (e.g., unstructured) format may be input in a large language model (LLM). The LLM may generate a first natural language summary associated with the data object based on the set of metadata. After receiving a natural language query from a user, the LLM may generate a second natural language summary associated with the data object based on the natural language query. The natural language summaries may be vectorized, and the vectorized versions may be compared. Based on the comparison, an indication of the data object corresponding to the natural language query may be displayed.