Context-Aware Apparatus Using T-Box and A-Box Segmentation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional reasoning devices lack intuition and struggle with accurately inferring user context, making knowledge-based reasoning complex and difficult, necessitating a technology for fast and easy context recognition using sensing information.
Innovation Solution
A context-aware apparatus and method that extracts a terminological-box (T-box) from a semantic model, generates a reasoning rule based on the T-box, and infers user context using an assertion-box (A-box) generated from sensing information, without relying on the T-box for inference, allowing for efficient merging and updating of A-boxes to improve context reasoning performance.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If knowledge-based reasoning is performed on user context, then reasoning accuracy is improved, but device complexity and difficulty increase
Solution Approach 1:
The semantic model is segmented into two distinct boxes: T-box (terminological box) containing class definitions and relationships, and A-box (assertion box) containing instance information and facts. This segmentation allows the reasoning system to separate static knowledge structures from dynamic sensing information, simplifying the reasoning process while maintaining accuracy.
Solution Approach 2:
The T-box is extracted from the semantic model as a separate component that defines relationships between classes. This extracted T-box is then used to generate reasoning rules that guide the inference process, allowing the system to perform accurate reasoning without carrying the entire semantic model structure, thus reducing complexity.
2Loss of information
If T-box is used for inference, then reasoning completeness is improved, but memory usage increases
Solution Approach 1:
The T-box is extracted from the semantic model and used specifically for generating reasoning rules, while the actual inference process operates on the smaller A-box containing only instance information. This extraction allows the system to maintain reasoning completeness by preserving the T-box structure for rule generation while reducing memory usage during the actual inference operation.
Solution Approach 2:
Reasoning rules are generated in advance from the T-box before the actual inference process. This preliminary action allows the system to prepare the reasoning framework beforehand, so that during runtime, only the A-box needs to be processed against these pre-generated rules, reducing the computational and memory burden during inference.
3Reliability
If conventional reasoning methods are used, then theoretical completeness is improved, but processing speed decreases
Solution Approach 1:
The reasoning process is segmented into distinct phases: T-box extraction for rule generation, A-box generation from sensing information, and inference execution. This segmentation allows each phase to be optimized independently, maintaining theoretical completeness while improving overall processing speed through efficient separation of knowledge representation and reasoning execution.
Solution Approach 2:
The T-box extraction and reasoning rule generation are performed as preliminary actions before actual context recognition. This allows the system to prepare the reasoning framework in advance, so that when sensing information arrives, the inference can be executed quickly by matching against pre-generated rules rather than building the reasoning structure in real-time.
Data Source
AI summary
A context aware apparatus is provided. The context aware apparatus includes an extracting unit configured to extract a terminological-box (T-box) from a semantic model, a first generating unit configured to generate a reasoning rule based on the extracted T-box, a second generating unit configured to generate a first assertion-box (A-box) based on sensing information, and a reasoning unit configured to infer a user context based on the reasoning rule and the first A-box.