Trajectory-Aware Mobile Search Prediction
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing portable devices and navigation systems struggle to provide relevant search results for users in transit, as they rely on current location and require pre-defined destinations, failing to offer useful suggestions when users are traveling and have not specified a route or destination.
Innovation Solution
The system and method for mobile computing devices to predict future locations based on location history, creating a trajectory-aware search query that includes search terms and geographic information, allowing for relevant search results to be generated even without a specified destination or route.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If search results are based on current location only, then the search system is simple and fast, but the search results are not relevant for users in transit who have already passed the location
Solution Approach 1:
The system performs preliminary actions by predicting future locations before the user actually arrives there. It analyzes current location, movement direction, and speed to proactively determine where the user is heading, then pre-prepares relevant search results for those predicted locations. This resolves the contradiction by making search results relevant in advance without requiring complex real-time analysis when the user searches.
Solution Approach 2:
The search system transitions from a static current-location-based model to a dynamic trajectory-based model. It continuously updates the predicted location based on changing movement parameters (speed, direction, current position), allowing the search results to dynamically adapt to the user's journey rather than being fixed to a single point. This dynamic approach improves relevance without requiring the entire system to be complex.
2Reliability
If the system requires users to specify destination and route, then search results can be tailored to the route, but the system becomes less convenient and requires more user input
Solution Approach 1:
The system performs self-service by automatically inferring the user's intended destination and route without requiring explicit user input. It uses the device's location data, movement patterns, and trajectory analysis to autonomously determine where the user is heading and what search results would be relevant. This resolves the contradiction by maintaining accurate, route-tailored search results while eliminating the need for users to manually specify destinations.
Solution Approach 2:
The system uses feedback from location sensors and movement data to continuously refine its prediction of the user's destination and route. By monitoring changes in position, speed, and direction over time, the system adapts its search results based on actual user behavior rather than relying on user-provided information. This feedback mechanism enables accurate search results with minimal user input.
3Ease of operation
If the system provides search results for locations the user has already passed, then the system is simple to implement, but the search results are not useful for traveling users
Solution Approach 1:
The system applies preliminary action by predicting future locations before the user reaches them and preparing relevant search results in advance. Instead of simply listing locations the user has already passed (simple but useless), it proactively identifies upcoming locations along the user's trajectory and prepares appropriate search results. This maintains relative implementation simplicity while dramatically improving usefulness.
Data Source
AI summary
A method includes receiving one or more search terms at a mobile computing device while the mobile computing device is located at a particular location. A search query that includes the one or more search terms and a location history of the mobile computing device is transmitted to a server. The method also includes receiving one or more search results in response to the search query, where the one or more search results include content identified based on a predicted destination of the mobile computing device. An interface identifying the one or more search results is displayed at the mobile computing device.