Medical Knowledge Discovery Using Semantic Association Sorting
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing methods for automatically discovering medical knowledge from large volumes of literature face challenges in distinguishing useful information and providing semantic explanations, often generating numerous irrelevant target concepts and lacking semantic resources for reasonable knowledge discovery.
Innovation Solution
A method and apparatus utilizing a pretrained Markov logic network to calculate association degrees between concepts based on semantic relations, filtering and sorting target concepts to enhance the discovery of useful medical knowledge, and providing logical explanations for the relations between starting and target concepts.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Quantity of substance
If statistical information-based methods are used to discover medical knowledge, then the quantity of processed literature can be handled, but the quality of discovered knowledge decreases due to generation of many useless target concepts
Solution Approach 1:
The patent introduces semantic resources (knowledge graphs, semantic networks) as intermediaries between statistical co-occurrence data and final knowledge discovery. These semantic structures mediate the relationship by providing predefined meaningful associations between medical concepts, allowing the system to process large volumes of literature while filtering results through semantic validity checks, thus maintaining high knowledge quality without sacrificing processing scale.
Solution Approach 2:
The patent transforms the approach by changing from purely statistical parameters (co-occurrence frequency) to semantic parameters (relationship types, concept hierarchies, semantic similarity scores). This parameter transformation enables the system to evaluate potential associations not just by frequency but by semantic meaningfulness, thereby reducing useless target concepts while maintaining ability to process large literature volumes.
2Loss of information
If semantic relations are introduced to explain discovered knowledge, then the explainability improves, but the ability to distinguish importance of knowledge decreases
Solution Approach 1:
The patent segments the evaluation of discovered knowledge into multiple independent dimensions: semantic relation type (for explainability), association strength score (for importance), and semantic resource validation (for quality). By dividing the evaluation process into these separate segments, the system can simultaneously provide detailed semantic explanations while also ranking and distinguishing the relative importance of different discovered associations through dedicated scoring mechanisms.
3Productivity
If no semantic resources are used in statistical methods, then the processing speed is fast, but the ability to provide reasonable explanations is lost
Solution Approach 1:
The patent performs preliminary action by pre-building and indexing semantic resources (knowledge graphs, concept hierarchies, relationship ontologies) before the actual literature processing begins. These semantic structures are constructed in advance and stored for rapid lookup during knowledge discovery, allowing the system to maintain fast processing speeds while still providing semantic explanations, as the semantic validation requires only quick queries to pre-computed structures rather than complex real-time analysis.
Data Source
AI summary
Embodiments of the present disclosure provide a method and an apparatus for automatically discovering medical knowledge. In this method, one or more linking concepts having a semantic relation with a starting concept are obtained from a medical literature library. The starting concept represents a disease. Next, one or more target concepts having a semantic relation with the one or more linking concepts are obtained from the medical literature library, and an association degree of each of the one or more target concepts with respect to the starting concept is calculated. The association degree indicates a probability that the target concept can cope with the starting concept. Further, the one or more target concepts are sorted according to the calculated association degrees. In this method, explainable target concepts can be obtained by using semantic analysis, and these target concepts are sorted to increase a possibility of discovering useful medical knowledge.