Routine Deviation Notification via Spatial Clustering
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Social-networking systems face challenges in accurately detecting deviations from a user's routine location, which limits their ability to send relevant notifications to connected users based on the user's current location relative to their expected routine.
Innovation Solution
The system estimates a user's routine by grouping ambient-location readings and calculates deviations from expected locations based on time of day and day of the week, using spatial-clustering algorithms to identify routine centers and send notifications to connected users when significant deviations occur.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If the system sends notifications based on location deviation detection, then social interaction and communication are enhanced, but the accuracy of detecting routine deviations is insufficient
Solution Approach 1:
The system performs preliminary actions by collecting and storing ambient location readings over time before actual deviation detection is needed. This historical data is processed to establish the user's routine pattern, including routine centers, confidence values, and expected locations at specific times. When deviation detection is required, the pre-established routine model enables accurate comparison without real-time computation delays, thereby improving both notification relevance and detection accuracy.
2Loss of time
If spatial-clustering algorithms are used to estimate user routine, then notification timing and relevance are improved, but computational complexity increases
Solution Approach 1:
The spatial-clustering algorithms are executed in advance during off-peak times to establish the user's routine pattern. The system processes historical location data to identify routine centers and confidence values beforehand, storing these results for rapid retrieval during deviation detection. This preliminary computation eliminates real-time processing delays, ensuring timely notifications while managing computational complexity through batch processing.
Solution Approach 2:
The system applies partial action by focusing computational resources on identifying only the most significant routine patterns and centers rather than analyzing every possible location variable. The confidence value metric allows the system to prioritize high-confidence routine patterns, reducing computational complexity while maintaining sufficient accuracy for effective deviation detection and timely notification delivery.
Data Source
AI summary
In one embodiment, a method includes determining a pattern among a number of geo-locations of a user over a period of time. Each geo-location is determined and reported by a mobile computing device of the user without manual user input. The method also includes determining a particular geo-location among the geo-locations in the pattern with a minimum distance to a current geo-location of the mobile computing device; determining a distance between the particular geo-location and the current geo-location; and sending a notification of the current geo-location to one or more other users in response to the distance being longer than a pre-determined deviation threshold value. One or more of the other users have a relationship to the user based at least in part on social-graph information associated with the user.