GPS Activity Inference via Location Database Cross-Reference
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing applications rely on self-reported data from users for tracking activities, which can be inaccurate due to user laziness, forgetfulness, or deceitfulness, and lack the ability to infer activities based on user location.
Innovation Solution
A method and system that infers user activities based on their location using GPS tracking, cross-referencing location records with a known-locations database to generate a list of inferred activity records, which can then be used to augment or replace self-reported data.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If self-reported data collection is used, then users can provide activity information, but data accuracy deteriorates due to user laziness, forgetfulness, or deceitfulness
Solution Approach 1:
The system uses automatically collected location data to generate activity records without requiring user input. The location data serves itself to infer activities, eliminating the need for users to manually report their activities while maintaining high data accuracy through objective tracking.
Solution Approach 2:
The patent replaces the manual mechanical process of self-reporting with an automated electronic system that uses GPS location tracking and database matching to infer activities. This substitution eliminates human error and dishonesty while continuously collecting accurate data.
2Measurement precision
If location tracking is implemented to automatically generate activity lists, then data accuracy improves, but device complexity increases
Solution Approach 1:
The system uses a universal location tracking mechanism (GPS) that serves multiple functions: determining user position, inferring activities, and generating activity records. By making the location data multi-functional, the system avoids adding separate complex detection mechanisms while achieving accurate activity detection.
Solution Approach 2:
The patent introduces a known-locations database as an intermediary between raw location data and activity inference. This database acts as a mediator that translates coordinate information into meaningful activity categories, simplifying the overall system architecture while maintaining high accuracy.
3Measurement precision
If continuous location monitoring is performed, then activity inference accuracy improves, but energy consumption increases
Solution Approach 1:
The system performs location monitoring periodically rather than continuously, checking location at intervals sufficient to detect activity changes while allowing the device to enter low-power states between checks. This periodic approach maintains activity detection accuracy while significantly reducing energy consumption.
Solution Approach 2:
The patent monitors location data with sufficient frequency to capture activity transitions, but not at the maximum possible rate. This partial monitoring approach provides enough data for accurate activity inference without the excessive energy consumption of continuous high-frequency tracking.
Data Source
AI summary
A method and system for influencing or forming a list of activities of a user from the location of the user. In one embodiment, a user is tracked with a GPS mobile tracker. The list of locations the user was present at is correlated with a known locations database to produce a list of user activities, with the type of activity designated by the name of the establishment at the location of the user. A further embodiment cross-references the establishment with a known activities database to filter out spurious records, where the user was not present for long enough or was present for too long for the record to be considered meaningful in an application. A further embodiment uses the location information to deduce information about the health of the user.