Route Search System Using Movement Path Classification
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing route search systems fail to accurately analyze the frequency of link usage based on user tastes, as they rely on pre-defined taste items that may not reflect individual user preferences, leading to inappropriate cost computations and route selections.
Innovation Solution
A system that classifies movement paths of users into types based on their characteristics, allowing for objective classification and cost setting for each type, enabling route searches that reflect the tastes of multiple users by setting costs for each type and attribute of the road.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Ease of manufacture
If the frequency of use of links is analyzed separately for each pre-defined taste item (such as age, vehicle type), then the analysis can be performed with a simple classification framework, but the analysis may be inappropriate if the user's taste does not depend on those pre-defined items
Solution Approach 1:
Instead of classifying users into pre-defined taste items and analyzing frequency of use for each item, the patent inverts the approach by directly analyzing the frequency of use of links themselves to determine user preferences. The system extracts user preferences from actual movement path data without requiring pre-defined classification categories, thereby achieving accurate preference analysis while avoiding the limitations of artificial classification frameworks
Solution Approach 2:
The patent changes the parameter of classification from pre-defined user attributes (age, vehicle type) to dynamically extracted movement path characteristics. By analyzing actual movement data to identify frequently used links and patterns, the system adapts the classification parameters to reflect real user behavior rather than relying on fixed demographic categories
2Adaptability or versatility
If costs are set for each type and each attribute of the road based on movement paths of multiple users, then the route search can reflect the tastes of multiple users, but the processing complexity increases
Solution Approach 1:
The patent merges the preference information of multiple users by analyzing aggregated movement path data. Instead of processing each user's preferences separately and then combining them, the system directly analyzes collective movement patterns to extract common preferences and set costs that reflect the tastes of multiple users simultaneously, thereby reducing processing complexity while maintaining adaptability
Solution Approach 2:
The patent uses movement path data from multiple users as copies of actual user behavior to infer preferences. By analyzing these copied behavioral patterns rather than requiring explicit preference inputs from each user, the system can efficiently derive cost parameters that reflect collective user tastes without the complexity of individualized preference elicitation and processing
Data Source
AI summary
Route search systems, methods, and programs acquire movement paths of a plurality of users, and classify, on the basis of a plurality of characteristics of the movement paths of the plurality of users, the movement paths of each of the users into a plurality of types such that movement paths with a plurality of similar characteristics belong to a same type. The systems, methods, and programs set a cost for each of the types and each attribute of a road on the basis of the plurality of movement paths of each of the types, classify, on the basis of a plurality of characteristics of a movement path of a route searcher, the route searcher into one of the types, and search for a route on the basis of the cost which is set for the type of the route searcher.