Geolocation Accuracy via User Interaction History
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing geolocation technologies are not precise enough to accurately determine the physical location of an electronic device within a building, often assigning the wrong address due to the influence of surrounding structures, leading to incorrect location data.
Innovation Solution
A method and system that calculates a most probable physical location of an electronic device by combining user-specific interaction history with statistical data from other users, using a computer device to receive geolocation data, identify probable locations, and compute user-specific and non-specific probability factors to select the most accurate location based on interaction history and type.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If traditional geolocation technology is used to determine physical location, then the system is simple and fast, but the measurement precision is insufficient and addresses are often incorrect
Solution Approach 1:
The patent segments the location determination process into multiple independent components: geolocation data reception, probable location identification, user interaction history establishment, probability factor calculation (both user-specific and user non-specific), and final address selection. Each component processes specific data types and produces intermediate results that feed into the next stage, allowing complex processing to be managed through modular, sequential operations
Solution Approach 2:
The patent adds a temporal dimension to location determination by incorporating user interaction history over time. Instead of relying solely on spatial geolocation data, the system analyzes historical interaction patterns across multiple time points, creating a time-based probability profile that complements the spatial information and significantly improves location accuracy
2Reliability
If geolocation data alone is used to determine location, then the processing is fast, but the reliability is low due to building structure interference
Solution Approach 1:
The patent performs preliminary actions by pre-establishing user interaction histories and probability factors before actual location determination is needed. The system continuously accumulates and processes interaction data in the background, so when location determination is required, the computationally intensive probability calculations are already complete or near-complete, reducing real-time processing requirements
Solution Approach 2:
The patent introduces probability factors as intermediary variables that mediate between raw geolocation data and final address determination. These probability factors (both user-specific and user non-specific) act as intermediate computational layers that filter and interpret geolocation data, accounting for building structure interference and other reliability issues before committing to a final location assignment
3Measurement precision
If only user-specific interaction history is used, then the results are personalized, but the statistical significance is reduced
Solution Approach 1:
The patent merges user-specific interaction history with user non-specific interaction history to create a comprehensive probability assessment. The user-specific component provides personalized accuracy by capturing individual behavior patterns, while the user non-specific component (aggregated from multiple users) provides statistical significance and robustness. The final probability factor combines both components, leveraging the strengths of each data source
Data Source
AI summary
A method of determining an address corresponding to a most probable physical location of an electronic device associated with a user is executable on a computer device and comprises receiving geolocation data from the electronic device. Based on received geolocation data, at least two probable physical locations of the electronic device will be found, with each of the at least two probable physical locations corresponding to a physical entity. Each physical entity is selected from a predetermined list and is associated with a physical entity type. A user interaction history is established, with respect to the at least two physical entities.