LLM Graph Querying for Reviewable Enterprise Data Outputs
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Enterprise organizations face challenges in efficiently extracting relevant information from large data graphs due to cumbersome query results and limited user interaction capabilities, especially when time is constrained.
Innovation Solution
Utilizing a large language model (LLM) to generate graph data queries and process data graph inputs, enabling efficient extraction and summarization of relevant information based on user requests, without requiring explicit user-defined prioritization.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Loss of information
If a data graph is used to store entity relationships, then information completeness is improved, but query complexity increases
Solution Approach 1:
The patent introduces a natural language processing intermediary layer that mediates between the user and the complex data graph query system. Users provide natural language requests, and the system automatically translates these into appropriate graph queries, eliminating the need for users to directly construct complex graph queries while preserving complete information retrieval from the data graph structure.
2Loss of information
If comprehensive query results are obtained from the data graph, then information completeness is improved, but result reviewability deteriorates
Solution Approach 1:
The patent segments the comprehensive query results into organized, digestible portions that maintain information completeness while improving reviewability. The system structures the retrieved data from the data graph in a hierarchical or categorized manner, allowing users to review information systematically rather than facing an overwhelming unstructured result set.
Solution Approach 2:
The natural language processing system acts as an intermediary that not only translates user requests into graph queries but also transforms the raw graph query results into user-friendly formats. This intermediary layer processes the comprehensive data and presents it in a structured, reviewable manner while preserving all relevant information.
3Loss of information
If more time is allocated for query execution, then information completeness is improved, but productivity deteriorates
Solution Approach 1:
The system performs preliminary actions by pre-processing and understanding the user's natural language request before executing the graph query. The natural language processing component analyzes the intent, identifies key entities and relationships, and formulates an optimized query plan in advance, allowing for faster execution while maintaining information completeness.
Solution Approach 2:
The patent implements dynamic query optimization where the system adapts query execution strategies based on the specific request and data characteristics. The natural language processing layer dynamically determines the most efficient query paths through the data graph, balancing information completeness with execution time to maintain productivity.
Data Source
AI summary
Systems and methods for generating output data based on a data graph are provided. An output request for the output data based on the data graph is received. The output request comprises one of a natural language request from a target user or an application request from an application of the target user. The data graph has nodes and edges between the nodes. The nodes represent entities associated with an enterprise organization and the edges represent relationships among the entities. A graph data query is generated with a large language model (LLM) using the output request as a first input to the LLM. The graph data query is performed against the data graph to obtain a graph data output that represents a sub-portion of the data graph. The output data is generated with the LLM using the graph data output as a second input to the LLM.