ONLINE MATCHING OPTIMIZATION DEVICE, METHOD, AND PROGRAM

The optimization device optimizes both the matching strategy and variables to control node appearance and edge reward, addressing the limitations of existing techniques in online matching.

JP7582500B2Active Publication Date: 2024-11-13NIPPON TELEGRAPH & TELEPHONE CORP
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
JP2023550853
Authority / Receiving Office
JP · JP
Patent Type
Patents
Current Assignee / Owner
Filing Date
2021-09-29
Publication Date
2024-11-13
Estimated Expiration
2041-09-29

AI Technical Summary

Technical Problem

Existing techniques for online matching in bipartite graphs fail to simultaneously optimize the matching strategy and variables that control the probability of node appearance and edge reward.

Method used

An optimization device and method that determines an optimal matching strategy and variables by formulating an optimization problem using parameter information, including appearance probability and edge rewards, and solving it to control both the probability of node appearance and edge reward.

Benefits of technology

Simultaneously optimizes the matching strategy and variables to control the probability of node appearance and edge reward, providing efficient allocation of resources in online services.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure 0007582500000012
    Figure 0007582500000012
  • Figure 0007582500000013
    Figure 0007582500000013
  • Figure 0007582500000014
    Figure 0007582500000014
Patent Text Reader

Abstract

One aspect of the present invention involves acquiring parameters including: a probability function that defines probabilities of appearance of a first node at a plurality of time points; a reward to be assigned when an edge is matched among a set of edges serving to associate sets of first nodes with sets of second nodes for the plurality of time points; and a period required until a second node corresponding to the matched edge can be used again for the plurality of time points. Then, one aspect of the present invention involves defining a first optimization problem formulated by using the acquired parameters, and solving the formulated first optimization problem to determine an optimal solution including variables for controlling the reward of an edge and the probability of appearance and a matching strategy for specifying a second node to be assigned to a first node that has appeared, and outputting the determined variables and matching strategy.
Need to check novelty before this filing date? Find Prior Art