《计算机应用》唯一官方网站 ›› 2025, Vol. 45 ›› Issue (6): 1934-1944.DOI: 10.11772/j.issn.1001-9081.2024060747

• 先进计算 • 上一篇    

空间众包中支持周期性任务的在线匹配算法

刘俊岭1,2, 孙萌1,2, 孙焕良1,2(), 许景科1,2,3   

  1. 1.沈阳建筑大学 计算机科学与工程学院,沈阳 110168
    2.辽宁省城市建设大数据管理与分析重点实验室(沈阳建筑大学),沈阳 110168
    3.国家特种计算机工程技术研究中心沈阳分中心(沈阳建筑大学),沈阳 110168
  • 收稿日期:2024-06-11 修回日期:2024-08-22 接受日期:2024-08-26 发布日期:2024-09-06 出版日期:2025-06-10
  • 通讯作者: 孙焕良
  • 作者简介:刘俊岭(1972—),女,辽宁沈阳人,副教授,博士,CCF会员,主要研究方向:时空数据查询、数据挖掘
    孙萌(2000—),女,山东莱阳人,硕士研究生,CCF会员,主要研究方向:时空众包、数据挖掘
    孙焕良(1969—),男,黑龙江望奎人,教授,博士生导师,博士,CCF高级会员,主要研究方向:空间数据管理、数据挖掘 sunhl@sjzu.edu.cn
    许景科(1976—),男,辽宁海城人,教授,博士,CCF会员,主要研究方向:时空数据库、数据挖掘。
  • 基金资助:
    国家重点研发计划项目(2021YFF0306303);辽宁省教育厅项目(LJKMZ20220916)

Online matching algorithm for supporting periodic tasks in spatial crowdsourcing

Junling LIU1,2, Meng SUN1,2, Huanliang SUN1,2(), Jingke XU1,2,3   

  1. 1.School of Computer Science and Engineering,Shenyang Jianzhu University,Shenyang Liaoning 110168,China
    2.Liaoning Province Big Data Management and Analysis Laboratory of Urban Construction (Shenyang Jianzhu University),Shenyang Liaoning 110168,China
    3.Shenyang Branch of National Special Computer Engineering Technology Research Center (Shenyang Jianzhu University),Shenyang Liaoning 110168,China
  • Received:2024-06-11 Revised:2024-08-22 Accepted:2024-08-26 Online:2024-09-06 Published:2025-06-10
  • Contact: Huanliang SUN
  • About author:LIU Junling, born in 1972, Ph. D., associate professor. Her research interests include spatio-temporal data query, data mining.
    SUN Meng, born in 2000, M. S. candidate. Her research interests include spatial crowdsourcing, data mining.
    SUN Huanliang, born in 1969, Ph. D., professor. His research interests include spatial data management, data mining.
    XU Jingke, born in 1976, Ph. D., professor. His research interests include spatio-temporal database, data mining.
  • Supported by:
    National Key Research and Development Program of China(2021YFF0306303);Project of Educational Department of Liaoning Province(LJKMZ20220916)

摘要:

针对空间众包中的一类重复多次且有固定需求的周期性任务,现有的匹配算法忽略了周期性任务对熟悉度的需求,提出一种空间众包中支持周期性任务的在线匹配算法。首先,将在线匹配问题视为多人游戏,将任务视为游戏的独立参与者,根据任务倾向于匹配熟悉度高的工人,工人倾向于匹配报酬高、距离近的任务的需求,确定玩家的效用函数,并进行博弈论(GT)分析;其次,将模拟退火(SA)策略引入GT的更新策略中,设计基于SA策略的GT算法;最后,在达到纳什均衡时实现总效用更大的匹配。在真实数据集上的实验结果表明,相较于现有相关算法,所提GT算法的匹配具有最高的总效用。可见,所提GT算法实现了更符合周期性任务和工人各自需求的匹配结果,可以提升在线空间众包平台的用户满意度。

关键词: 空间众包, 周期性任务, 博弈论, 总效用, 熟悉度

Abstract:

For a type of repetitive periodic tasks with fixed demand in spatial crowdsourcing, the existing matching algorithms ignore the familiarity required for periodic tasks, so that an online matching algorithm for supporting periodic tasks in spatial crowdsourcing was proposed. Firstly, the online matching problem was regarded as a multiplayer game, with tasks considered as independent participants in the game, and utility functions of the players were determined based on the need of tasks’ preference for matching workers with high familiarity and workers’ preference for tasks with high rewards and short distances, which were then analyzed using Game Theory (GT). Then, a Simulated Annealing (SA) strategy was introduced into the updating strategy of GT, resulting in the design of GT algorithm based on SA strategy. Finally, a matching with greater total utility was achieved with reaching a Nash equilibrium. Experimental results on real datasets demonstrate that the proposed GT algorithm achieves the highest total utility compared to existing related algorithm, the matching of the proposed GT algorithm has the highest total utility. It can be seen that the proposed GT algorithm achieves matching results that better meet the need of periodic tasks and workers, which can enhance user satisfaction on online spatial crowdsourcing platforms.

Key words: spatial crowdsourcing, periodic task, Game Theory (GT), total utility, familiarity

中图分类号: