Taxonomy-Based Geofence Discovery and Annotation System
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current methods for discovering and annotating functional areas in cities are inefficient, as they separate the formation of functional areas and their semantic annotation, leading to unclear and labor-intensive descriptions, and existing approaches rely on topic-based inference models or segmentation techniques that fail to provide a clear understanding of area meaning.
Innovation Solution
A taxonomy-based system that divides an area into cells, assigns initial labels, and applies hierarchical clustering to find clusters with common labels, using an objective function that balances spatial adjacency and label generalization, allowing for efficient discovery and annotation of functional clusters.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If topic-based inference models or segmentation techniques are used for area annotation, then annotation can be performed, but the annotation is labor-intensive and fails to provide clear understanding of area meaning
Solution Approach 1:
The patent combines spatial clustering and semantic annotation into a single integrated process. The hierarchical clustering algorithm simultaneously groups geo-referenced points into functional areas and assigns semantic labels, eliminating the need for separate annotation steps and reducing labor intensity while maintaining clear area meaning through the dual objective function.
Solution Approach 2:
The system performs self-annotation by automatically assigning semantic labels to discovered functional areas through the clustering process. The objective function inherently guides the annotation by favoring labels that clearly distinguish area functions, enabling the system to annotate areas without manual intervention while preserving meaningful semantic information.
2Productivity
If spatial clustering techniques are used to discover functional areas, then area discovery can be performed, but the derived areas may not be clearly annotated
Solution Approach 1:
The patent merges the area discovery and annotation processes into a unified hierarchical clustering framework. The algorithm simultaneously performs spatial grouping and semantic labeling in one operation, ensuring that discovered areas are inherently well-annotated through the objective function that optimizes both spatial coherence and label clarity.
Solution Approach 2:
The system uses an objective function with adjustable parameters to control the balance between spatial clustering quality and annotation clarity. By optimizing this parameterized function, the system discovers functional areas with clear semantic meanings, transforming the trade-off into a controllable optimization problem.
3Measurement precision
If hierarchical clustering is applied with objective function maximization, then clusters with common labels are found, but computational complexity increases
Solution Approach 1:
The patent applies hierarchical clustering that segments the area discovery process into manageable levels of granularity. By organizing clustering operations hierarchically and using an objective function to guide label assignment at each level, the system achieves precise cluster annotation while breaking down computational complexity into smaller, more manageable sub-problems.
Data Source
AI summary
Systems and methods for discovering and annotating geo-fences from geo-referenced data are disclosed. The systems and methods input an area of interest containing a plurality of geo-referenced points having associated labels, and divides the area interest into cells. Each cell is assigned an initial label from among the plurality of labels and hierarchical clustering is used to find clusters of cells having a common label based on a maximization of an objective function for each cell with the objective function being dependent upon favoring spatially adjacent cells having a common label and limiting overgeneralization of the common label.