Semantic Reasoning for Personal Illness Management
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current semantic networks lack mechanisms for automatic and systematic addition of new content through inference, and they do not provide a way to represent history, limiting their ability to learn and adapt over time.
Innovation Solution
A computer system and method that incorporates an extended semantic model with an inference engine for personal illness management, which automatically derives insights by adding new nodes and links to a semantic knowledge database using inference logic, enabling the system to learn and adapt over time.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Extent of automation
If knowledge is explicitly added to the semantic network through direct association, then the knowledge base is populated with facts, but the system lacks automatic reasoning capability to derive new knowledge
Solution Approach 1:
An inference engine is introduced as an intermediary component between the semantic network and external data sources. This engine applies reasoning rules to existing knowledge and automatically derives new nodes and links, enabling automatic knowledge derivation without requiring manual addition of every fact while maintaining a manageable system structure through separation of reasoning logic and knowledge storage
Solution Approach 2:
The system pre-defines a comprehensive set of reasoning rules and inference patterns within the inference engine before operation. These preliminary-configured rules enable the system to automatically derive new knowledge from existing facts without requiring real-time complex decision-making about what knowledge to add, thus automating the knowledge derivation process while controlling system complexity through advance preparation
2Adaptability or versatility
If the semantic network stores only current knowledge states, then the structure remains simple, but the system cannot represent historical data or learn from changes over time
Solution Approach 1:
The system adds a temporal dimension to the semantic network by incorporating time-stamped versions of nodes and links that represent knowledge states at different points in time. This allows the network to store historical data and evolve over time while maintaining the ability to query and reason about past states, thus gaining temporal reasoning capability without overwhelming data volume through selective historical retention
Solution Approach 2:
The semantic network is designed as a dynamic structure where nodes and links can be added, modified, or deleted over time to reflect changing knowledge states. The inference engine continuously applies reasoning rules to update the network, enabling the system to adapt and learn from changes while managing data volume through incremental updates rather than storing all intermediate states
3Productivity
If manual addition of nodes and links is used, then the semantic network can be constructed, but the process is time-consuming and cannot keep pace with rapidly changing data
Solution Approach 1:
The inference engine is designed to autonomously apply reasoning rules to the semantic network without requiring manual intervention. It automatically derives new knowledge, adds new nodes and links, and updates the knowledge base based on existing facts and predefined rules, thus achieving rapid knowledge updates that keep pace with changing data while eliminating time-consuming manual processes
Data Source
AI summary
A personal illness management system includes an extended semantic model of a health care knowledge domain, a semantic knowledge database for personal illness management, and an inference engine. The extended semantic model of a health care knowledge domain includes existing concepts related to personal illness management, existing relationships among the existing concepts, and inference logic embedded within each existing concept. The semantic knowledge database for personal illness management is distinct from the extended semantic model and includes existing nodes and existing links. The existing nodes represent instances of the existing concepts, and the existing links represent instances of the existing relationships among the existing concepts. The inference engine is knowledge domain independent and populates the semantic knowledge database with the instances of the existing concepts and the instances of the existing relationships by following the inference logic embedded within each existing concept.