Semantic Reasoning for Learning Target Extraction in Education Metaverse
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current semantic reasoning methods for learning targets in education metaverse lack consideration for adjacent knowledge points, leading to inaccurate extraction and require manual association of teaching resources, making it difficult to organize and aggregate multi-source teaching resources effectively.
Innovation Solution
A semantic reasoning method that receives a description text, segments it into lexical items, extracts semantic items, maps them into word vectors, and determines association relationships between knowledge points, integrating these into a teaching model to automatically match teaching scenes and establish interaction relationships between models.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Extent of automation
If semantic reasoning is used to infer learning targets from knowledge graphs, then learning target inference capability is improved, but extraction accuracy deteriorates due to lack of adjacent knowledge points consideration
Solution Approach 1:
The patent segments the learning target extraction process into multiple independent modules: knowledge graph construction module, semantic reasoning module, adjacent knowledge point identification module, and resource matching module. Each module handles a specific aspect of the extraction process, allowing the system to maintain automated inference while improving accuracy through structured multi-step processing.
Solution Approach 2:
The patent introduces an intermediary knowledge graph structure that connects learning targets with adjacent knowledge points. This intermediary structure serves as a bridge between the semantic reasoning process and the final extraction results, enabling the system to consider contextual relationships and improve extraction accuracy without losing automation.
2Measurement precision
If manual marking is used for association relationships between teaching resources, then association accuracy is improved, but operation complexity and time consumption deteriorate
Solution Approach 1:
The patent implements self-service through automated semantic reasoning algorithms that automatically establish association relationships between teaching resources and learning targets. The system uses vector similarity calculations and knowledge graph reasoning to autonomously determine associations without requiring manual intervention, thereby reducing operational complexity while maintaining acceptable accuracy through algorithmic precision.
Solution Approach 2:
The patent incorporates feedback mechanisms where the system continuously refines association relationships based on matching results and resource usage data. The automated system learns from feedback loops, adjusting its association algorithms to improve accuracy over time without increasing manual operation complexity.
3Measurement precision
If manual editing is used for interaction behaviors between teaching resources, then interaction precision is improved, but productivity deteriorates due to tedious and complex processes
Solution Approach 1:
The patent enables self-service for interaction behavior generation through automated algorithms that analyze teaching resource characteristics and automatically generate appropriate interaction patterns. The system uses semantic reasoning to determine how resources should interact based on their relationships to learning targets, eliminating tedious manual editing while maintaining precision through algorithmic consistency.
Solution Approach 2:
The patent performs preliminary action by pre-defining interaction behavior templates and patterns that are automatically applied to teaching resources. Instead of manually editing each interaction, the system prepares interaction frameworks in advance and automatically populates them based on resource characteristics, significantly improving productivity while maintaining precision through template-based consistency.
4Productivity
If existing teaching resource library is used for generating teaching resources, then resource generation speed is improved, but adaptability deteriorates due to difficulty in organizing multi-source resources
Solution Approach 1:
The patent implements universality through a unified knowledge graph structure that can accommodate multiple types of teaching resources from different sources. The system uses a universal semantic reasoning framework that handles diverse resource formats and structures, enabling the system to maintain high generation speed while effectively integrating multi-source resources through a common processing architecture.
Data Source
AI summary
Disclosed are a semantic reasoning method and a terminal for a learning target in an education metaverse. The method includes receiving a description text of a learning target, segmenting the description text into a lexical item sequence, extracting an entry from the lexical item sequence, extracting a semantic lexical item based on the entry, and mapping the semantic lexical item into a first word vector. By capturing the semantic feature vector in the word vector, the subject and knowledge points in the learning target are obtained by reasoning, the target knowledge points of the learning target are determined and positioned, and the association relationship between the target knowledge points is obtained, so that the semantic information of the learning target can be fully mined, and the accuracy of the learning target extraction can be improved. It matches with the corresponding teaching scene based on the learning targets.