Individual Information Merging With Time-Aware Reliability Models
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing information fusion methods struggle with high false positives and require significant human intervention to distinguish between duplicate and distinct individuals due to the lack of consideration of time variance in property reliability, especially when comparing data from different sources.
Innovation Solution
An information processing method using ontology matching and evolution models to assess the reliability of properties over time, incorporating similarity distance calculations to merge instances of individuals effectively, reducing the need for human intervention by accounting for property variability.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If strict comparison between properties of individuals is performed independently of time difference, then comparison simplicity is maintained, but false positives increase and human intervention is required
Solution Approach 1:
The patent applies dynamics by making the comparison process adaptive to time differences. The system dynamically adjusts the comparison strategy based on the temporal gap between observations, transitioning from static property matching to time-aware evaluation. This resolves the contradiction by enabling automated time-sensitive comparison that maintains accuracy without requiring human intervention.
Solution Approach 2:
The patent changes the parameter of time difference from an irrelevant factor to a critical evaluation criterion. By incorporating time difference as a parameter that influences property reliability assessment, the system achieves more accurate identification while maintaining full automation. The evolution model adjusts property weights based on temporal factors, resolving the contradiction between accuracy and automation.
2Reliability
If time difference is not considered in property comparison, then processing speed is maintained, but reliability of identification decreases
Solution Approach 1:
The patent applies preliminary action by pre-defining evolution models for each property type that encode temporal degradation patterns. These models are prepared in advance and automatically applied during comparison, eliminating the need for complex real-time temporal analysis. This resolves the contradiction by achieving high reliability through pre-computed temporal adjustments without adding processing overhead.
Solution Approach 2:
The patent introduces an intermediary mechanism in the form of evolution models that mediate between raw property values and comparison results. These models act as temporal filters that automatically adjust property reliability based on time difference, enabling reliable identification without requiring time-consuming manual evaluation or complex real-time calculations.
3Measurement precision
If evolution models are applied to assess property reliability over time, then false positives are reduced, but system complexity increases
Solution Approach 1:
The patent applies segmentation by dividing the complexity into two parts: pre-defined evolution models for each property type and a simple application layer that uses these models during comparison. The evolution models encapsulate complex temporal reasoning separately, while the main system only needs to invoke them. This resolves the contradiction by achieving high accuracy through modular design without increasing overall system complexity.
Solution Approach 2:
The patent reduces system complexity by performing complex temporal analysis in advance through pre-defined evolution models. During runtime, the system simply applies these pre-computed models to property comparisons, avoiding the need for complex real-time calculations. This resolves the contradiction between accuracy and complexity by shifting computational burden to the model definition phase.
Data Source
AI summary
The method and system for merging information aimed at merging the instances of individuals, a data-processing system performs the following steps: generating the instances of individuals using an ontology which defines, for each property of each instance of an individual, an evolution model to be applied to the property, evolution model representing the evolution of reliability of the property over time in relation to variability of the property over time; preforming the merging of information by comparing, two-by-two, the generated instances of individuals with instances of individuals stored in a knowledge base, performing, for each shared property, a calculation of similarity distance by applying at least evolution model defined for the property, so as to define a coefficient of confidence for each property in order to decide whether or not to merge the instances of individuals; and updating the knowledge base with the instances of individuals resulting from information merging.