Pathfinding system, method and program

Abstract

The invention provides an art that makes defining multiple types that accurately reflect user preferences feasible. Configuring a pathfinding system provided with a movement path obtaining means that obtains movement paths for multiple users, a sorting means that, on the basis of multiple characteristics of the movement paths of the multiple users, sorts the movement paths of each of the users into multiple types such that the movement paths for which the multiple characteristics are similar are in the same type, a cost setting means that sets a cost for each of the types and each route attribute on the basis of the multiple movement paths in each type, a pathfinding person sorting means that, on the basis of multiple characteristics of a movement path of a pathfinding person, sorts the pathfinding person into any of the types, and a pathfinding means that finds a path on the basis of the cost set for the type of the pathfinding person.

Description

technical field [0001] The present invention relates to a route search system, method and program. Background technique [0002] Conventionally, techniques have been developed for enabling route search in accordance with the user's route selection preference. For example, Patent Document 1 discloses a method to calculate the frequency of use of a link as a preference degree according to preference items such as vehicle type, age, and gender, which may become important factors in route selection, and record it as a preference parameter. The attribute data similar to the preference parameters adjust the cost to search for the structure of the path (Patent Document 1, Claim 1, Figure 4 Wait). [0003] Patent Document 1: Japanese Patent Laid-Open No. 2009-42051 [0004] In the prior art described above, the frequency of use of links is analyzed according to predetermined preference items. In the prior art, these preference items are items that are artificially determined in ...

AI Technical Summary

Problems solved by technology

For example, even if the link usage frequency is analyzed by age, it cannot be properly analyzed if the link selection preference does not depend on age

In addition, since users who are younger than their actual age are considered (Patent Document 1,0051), even if corrections are made with reference to the frequency of use related to other ages, if the preference for link selection does not depend on age in the first place, the above-mentioned age-related The frequency of utilization may not be appropriate, so there is no point in correcting it

In this way, in the prior art, there may be cases where the analysis of the frequency of use that is the basis of the cost calculation becomes inappropriate

Images