Decision-Graph Detection of Minimal Inconsistent Event Sets
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing abduction methods struggle to reliably detect inconsistencies between observed events and knowledge bases, leading to incorrect hypothesis derivation.
Innovation Solution
A minimal unsatisfiable set detection apparatus and method that utilizes a decision graph to identify and refine inconsistencies by converting observed events into a minimal set covering problem, enumerating partial observation events directly linked to the inconsistency.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If conventional abduction methods are used to detect inconsistencies, then the process is simple, but the reliability of detection results is low
Solution Approach 1:
The patent segments the inconsistency detection process into distinct functional modules: a decision graph acquisition unit that constructs the knowledge representation, an observed event acquisition unit that collects input data, a minimal set covering derivation unit that performs the core analysis, and an output unit that presents results. This modular segmentation improves reliability by ensuring each component performs its function correctly while maintaining overall system coherence.
Solution Approach 2:
The patent introduces a decision graph as an intermediary data structure between the observed events and the inconsistency detection logic. This decision graph serves as a mediator that transforms raw observed events into a structured format that can be systematically analyzed for inconsistencies, thereby improving detection reliability without requiring direct complex processing of raw data.
2Measurement precision
If conventional inconsistency detection methods are used, then the process is fast, but the precision of detection results is low
Solution Approach 1:
The patent performs preliminary actions by constructing a decision graph from the knowledge base before inconsistency detection begins. This pre-processing step organizes the knowledge into a structured format with defined nodes and edges, enabling faster and more precise inconsistency detection during the actual analysis phase without requiring complex processing during the detection itself.
Solution Approach 2:
The patent applies partial action by focusing the inconsistency detection process specifically on the minimal set of observed events that cause inconsistencies, rather than analyzing all possible event combinations. The minimal set covering derivation unit identifies and analyzes only the essential subsets of observed events that lead to contradictions, improving precision while reducing the time required compared to exhaustive analysis.
3Measurement precision
If all observed events are analyzed for inconsistency, then completeness is high, but the ability to identify minimal inconsistent subsets is low
Solution Approach 1:
The patent extracts the minimal inconsistent subsets from the complete set of observed events through the minimal set covering derivation unit. This extraction process identifies and isolates the specific subsets of observed events that cause inconsistencies, separating them from the larger set of consistent events. This enables precise identification of minimal inconsistent subsets while maintaining analysis efficiency by focusing computational resources only on the relevant subsets.
Solution Approach 2:
The patent applies partial action by analyzing only the necessary subsets of observed events rather than all possible combinations. The minimal set covering derivation unit systematically identifies minimal subsets that cover all inconsistencies, performing exactly the analysis needed without redundant computations on supersets or unnecessary event combinations, thereby improving both precision and productivity.
Data Source
AI summary
A minimal unsatisfiable set detection apparatus, which detects a minimal set from a set of observed events that are determined to be inconsistent with a knowledge base, includes a decision graph acquisition unit 31 that acquires a decision graph that outputs a boolean value of an observed event from the knowledge base, an observed event acquisition unit 32 that acquires a plurality of observed events determined to be inconsistent with the knowledge base, a minimal set covering derivation unit 33 that obtains a minimal set covering by converting a plurality of observed events, which are acquired by the observed event acquisition unit 32, into a minimal set covering problem using the decision graph, and an output unit 34 that outputs a calculation result by the minimal set covering derivation unit 33.