Enterprise Semantic Models for Domain-Specific Knowledge Retrieval
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing data access systems in enterprise computing environments are inefficient for identifying domain-specific contextual information, leading to low reusability of knowledge assets, increased rework, and poor customer experience due to the lack of domain and industry-specific ontologies, complex folder structures, and high-priced functional-specific tools that do not cater to organizations like IT services, analytics services, and management consulting.
Innovation Solution
An artificial intelligence-based system that generates domain-specific semantic models using machine learning to identify and recommend storyboards for creating new knowledge assets, integrating with enterprise computing environments to enhance discoverability and relevance of knowledge assets.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Adaptability or versatility
If generic enterprise search tools are used to search for relevant information across all documentation tools, then broad coverage of information sources is achieved, but domain-specific contextual understanding and search precision deteriorate
Solution Approach 1:
The system segments the monolithic generic search tool into multiple domain-specific semantic models, each trained on specific domain data (e.g., IT services, analytics, consulting). This segmentation allows each model to specialize in its domain while the system as a whole maintains broad coverage across multiple domains.
Solution Approach 2:
The system applies local quality by making each semantic model domain-specific rather than uniformly generic. Each model has specialized knowledge and ontologies tailored to its specific domain, improving search precision locally within that domain while maintaining enterprise-wide versatility.
2Measurement precision
If functional-specific tools are used for documentation and discovery, then domain expertise and precision are improved, but cost and accessibility deteriorate
Solution Approach 1:
The system creates a universal enterprise search platform that incorporates multiple domain-specific semantic models. This allows a single system to serve multiple functions and domains (IT services, analytics, consulting, etc.) that would otherwise require separate expensive functional-specific tools, making domain expertise accessible enterprise-wide.
Solution Approach 2:
The system copies and adapts the specialized capabilities of expensive functional-specific tools into a unified platform. By training semantic models on domain-specific data from various sources, the system replicates domain expertise without requiring licensing multiple separate tools.
3Ease of manufacture
If knowledge assets are stored in complex folder structures, then organization and categorization are improved, but discoverability and access efficiency deteriorate
Solution Approach 1:
The system introduces semantic models as intermediaries between the complex folder structure and the user. Instead of requiring users to navigate complex folder hierarchies, the semantic models understand domain context and directly retrieve relevant knowledge assets, bypassing the need for manual folder navigation while maintaining the organizational structure.
Solution Approach 2:
The system replaces the mechanical navigation through folder structures with intelligent semantic search. Instead of manually traversing folders (mechanical process), users query the system which uses AI-driven semantic understanding to locate assets, substituting manual mechanical navigation with automated intelligent retrieval.
4Productivity
If employees focus on billable tasks to maximize revenue, then revenue generation is improved, but knowledge documentation and organization deteriorate
Solution Approach 1:
The system enables self-service knowledge retrieval where employees can independently find needed knowledge assets without requiring time for manual documentation or organization. The AI system automatically processes and organizes knowledge assets, allowing employees to focus on billable work while the system handles knowledge management autonomously.
Data Source
AI summary
The present disclosure provides a system and method for identifying domain specific contextual information with an enterprise computing environment. The system is configured to generate a domain specific semantic model (M) based on one or more domain specific articles and assets dataset from web, (ii) generate one or more enterprise specific semantic models for all N organizations by fine tuning the domain specific semantic model with the one or more articles and assets from an enterprise, and (iii) identifying a list of relevant knowledge assets within the enterprise using the one or more enterprise specific semantic models in response to a search query.