Knowledge Graph Querying for Cross-Project AI Knowledge Reuse
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing generative AI tools are limited in their ability to utilize insights from individual projects at an organizational scale, making it difficult to leverage knowledge across multiple projects within a corporate setting.
Innovation Solution
A natural text-based querying framework that utilizes a knowledge representation system to create individual work-record entities and knowledge graphs, performing operations such as name, subgraph, sentence embedding, and word embedding to facilitate intelligent re-use of knowledge across an organization.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Adaptability or versatility
If generative AI tools are used for individual projects, then task-specific capabilities are improved, but organizational-scale knowledge utilization deteriorates
Solution Approach 1:
The patent segments organizational knowledge into individual work-record entities, each representing a discrete project or task. Each entity is further divided into sub-entities (tasks, subtasks, actions, objects) that can be independently processed and reused. This segmentation allows task-specific AI tools to operate on granular units while maintaining connections to the broader organizational knowledge base.
Solution Approach 2:
The patent introduces an intermediary layer (the knowledge representation system with work-record entities and subgraph information) between individual project insights and organizational-scale utilization. This intermediary structures and indexes knowledge in a way that enables both specific task processing and broad organizational retrieval, bridging the gap between project-level and organizational-level knowledge use.
2Reliability
If knowledge is stored in individual project contexts, then project-specific insights are preserved, but organizational-scale re-use deteriorates
Solution Approach 1:
The patent merges individual project knowledge into a unified organizational knowledge base by creating work-record entities that capture project-specific details while establishing connections to other related entities. The subgraph information merges multiple aspects of work records (tasks, actions, objects) into an integrated structure that preserves project context while enabling organizational-scale retrieval and reuse.
Solution Approach 2:
The patent creates a universal knowledge representation framework that serves multiple functions: preserving project-specific insights, enabling organizational-scale retrieval, supporting various query types, and facilitating knowledge reuse across different contexts. The work-record entities and subgraph structures are designed to be universally applicable across diverse projects and organizational needs.
3Measurement precision
If detailed work record data is processed, then knowledge accuracy is improved, but processing complexity deteriorates
Solution Approach 1:
The patent segments detailed work record data into hierarchical levels (work-record entities, sub-entities, subgraphs) that can be processed at appropriate granularities. This segmentation reduces processing complexity by allowing operations to work on smaller, manageable units while maintaining the ability to reconstruct complete knowledge representations when needed for high accuracy.
Solution Approach 2:
The patent performs preliminary processing of work record data during the knowledge ingestion phase, organizing detailed information into structured work-record entities and subgraph representations before they are stored. This preliminary structuring reduces the complexity of subsequent retrieval and query operations while preserving the accuracy of the original detailed data.
Data Source
AI summary
Aspects of the present disclosure provide systems, methods, and computer-readable storage media that support intelligent re-use of knowledge (e.g., across an organization) using a natural text-based querying framework. A knowledge representation of prior work performed for the organization may be generated based on organizational knowledge (e.g., historical work record data that identifies a plurality of work items across an organization). The knowledge representation may include individual work-record entities for each respective work item and individual knowledge graphs corresponding to the individual work-record entities. For each individual knowledge graph, operations may be performed to identity and store project name, subgraph, sentence embedding, and word embedding information. Responsive to receiving an augmented user query, query-record mapping operations may be performed based at least in part on a comparison of information associated with the augmented user query to the project name, sentence embedding, word embedding, subgraph information, or a combination thereof.