Enterprise Knowledge Graphs for Accurate Generative AI Prompts
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing generative AI systems struggle with understanding buried concepts within data, leading to inaccurate and unreliable responses, especially in enterprise environments where user-specific information is dynamic and confidential, necessitating frequent and costly retraining.
Innovation Solution
Generating user-specific knowledge graphs (KGs) from enterprise data and using them as prompts for large language models (LLMs) to enhance understanding of relationships between concepts and entities, thereby improving accuracy and reliability without the need for frequent retraining.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If generative AI systems are trained on enterprise data to improve response accuracy, then the reliability of responses improves, but the training cost increases and information leakage risk increases
Solution Approach 1:
The patent segments enterprise data into user-specific knowledge graphs, organizing information by user context rather than training the entire model on all enterprise data. This allows the system to retrieve relevant knowledge on-demand without the computational cost of retraining on full datasets, maintaining response accuracy while reducing training expenses.
Solution Approach 2:
The patent introduces knowledge graphs as an intermediary between enterprise data and the generative AI model. Instead of directly training the model on raw enterprise data, the system converts data into structured knowledge graphs that serve as context for prompting, eliminating the need for frequent retraining while preserving response reliability.
2Reliability
If generative AI systems are trained on enterprise data to improve response accuracy, then the reliability of responses improves, but the risk of information leakage increases
Solution Approach 1:
The patent segments enterprise data into user-specific knowledge graphs, organizing information by user context rather than training the entire model on all enterprise data. This allows the system to retrieve relevant knowledge on-demand without the computational cost of retraining on full datasets, maintaining response accuracy while reducing training expenses.
Solution Approach 2:
The patent introduces knowledge graphs as an intermediary between enterprise data and the generative AI model. Instead of directly training the model on raw enterprise data, the system converts data into structured knowledge graphs that serve as context for prompting, eliminating the need for frequent retraining while preserving response reliability.
3Reliability
If generative AI systems use general enterprise data to improve response accuracy, then the reliability improves, but the understanding of user-specific concepts and relationships deteriorates
Solution Approach 1:
The patent applies local quality by creating user-specific knowledge graphs that tailor the structure and content of knowledge representation to each user's context. Instead of using a uniform enterprise-wide data structure, the system adapts the knowledge graph to reflect individual user relationships, concepts, and data access patterns, thereby improving concept understanding while maintaining overall system reliability.
Data Source
Figure 1
Figure 2A
Figure 2B
AI summary
A user-specific knowledge-based graph (KG) is generated from enterprise data corresponding to a user. The user-specific KG is stored in a data store. A query processing system receives a user query that is to be used in prompting a large language model (LLM) in an LLM service. The query processing system identifies portions of the KGs in the data store that relate to the query. A prompt generator generates a prompt to the LLM service using the identified portions of the KG and the query. A response processor receives a response from the LLM service and generates a response to the query based on the response received from the LLM service.