Disambiguating Clustered Location Identifiers Using Spatiotemporal Confidence
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
In high-density areas, accurately determining the location of captured digital assets is challenging due to imprecise GPS and location metadata, leading to difficulties in disambiguating clustered location identifiers and linking relevant assets.
Innovation Solution
The techniques involve generating a consolidated address for digital assets by determining geographic and time metadata, calculating confidence metrics based on distance and time ranges, and updating a knowledge graph to associate digital assets with precise location identifiers, using web services and location history to enhance accuracy.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If GPS and location metadata are used to determine capture location, then location information is obtained, but precision deteriorates in high-density areas with numerous establishments
Solution Approach 1:
The patent segments the location identification problem into multiple components: geographic location from metadata, time of capture, distance calculations to multiple establishments, and confidence metric evaluation. This segmentation allows systematic disambiguation of clustered location identifiers by evaluating each component separately and combining results.
Solution Approach 2:
The patent changes parameters by calculating distance ranges and time ranges, then using these parameter changes to compute confidence metrics. By transforming the raw metadata into derived parameters (distance, time, confidence scores), the system resolves ambiguity among clustered location identifiers with improved precision.
2Reliability
If manual labeling of digital assets is performed, then accurate location association is achieved, but time consumption increases significantly
Solution Approach 1:
The system performs self-service by automatically computing confidence metrics and associating digital assets with location identifiers without requiring manual user intervention. The algorithm independently evaluates distance ranges, time ranges, and confidence metrics to link assets to the most probable location, eliminating time-consuming manual labeling while maintaining high accuracy.
Solution Approach 2:
The patent implements feedback through confidence metrics that evaluate the likelihood of correct location association. This feedback mechanism allows the system to automatically adjust and refine location assignments based on calculated confidence levels, achieving reliable association without manual verification.
3Loss of information
If location metadata with ±200 meters precision is used for indoor locations, then location data is obtained, but difficulty increases in disambiguating clustered identifiers in high-density areas
Solution Approach 1:
The patent adds another dimension by incorporating time information alongside spatial coordinates. By evaluating both distance ranges and time ranges, the system creates a spatiotemporal framework that resolves ambiguity among clustered location identifiers, compensating for the limited spatial precision of indoor location metadata.
Solution Approach 2:
The system creates a composite location identification approach by combining multiple data sources: geographic coordinates, time metadata, distance calculations, and confidence metrics. This composite methodology integrates various information types to achieve accurate location disambiguation despite the inherent imprecision of individual metadata sources.
4Productivity
If digital assets are captured in close vicinity or shortly before/after visiting a location, then assets can be categorized with the visit, but accuracy decreases in determining actual capture location
Solution Approach 1:
The patent applies partial action by calculating confidence metrics that evaluate the degree of match between captured assets and potential locations. Rather than making binary assignments, the system computes partial confidence scores based on distance and time ranges, allowing accurate differentiation even when assets were captured in close proximity or near the location.
Data Source
AI summary
Embodiments of the present disclosure present devices, methods, and computer readable medium for disambiguating clustered location identifiers. A location identifier can be the name of a business or establishment. Digital assets contain a plurality of metadata that can be used to identify the location or establishment at which digital assets were captured. Techniques can use various types contextual information based on a category of the digital asset for disambiguation. Automatically labelling the digital assets assists a user in organizing and sharing the digital assets with friends and family. Users can search for digital assets by the name of the location where the digital assets were captured.