电子学报 ›› 2021, Vol. 49 ›› Issue (10): 2012-2019.DOI: 10.12263/DZXB.20200218

• 学术论文 • 上一篇    下一篇

移动群智感知中基于协同排序的任务推荐方法

王健1, 刘嘉欣1, 赵国生2, 赵中楠1   

  1. 1.哈尔滨理工大学计算机科学与技术学院,黑龙江 哈尔滨 150080
    2.哈尔滨师范大学计算机科学与信息工程学院,黑龙江 哈尔滨 150025
  • 收稿日期:2020-02-28 修回日期:2020-09-28 出版日期:2021-11-29
    • 作者简介:
    • 王 健 女,1979年6月出生,黑龙江哈尔滨人.毕业于哈尔滨工程大学,获得博士学位.现为哈尔滨理工大学教授、博士生导师,主要从事未来网络与认知网络、群智感知、可信计算等方面的研究工作. E-mail:wangjianlydia@163.com
      刘嘉欣 女,1996年9月出生,黑龙江哈尔滨人.现为哈尔滨理工大学计算机科学与技术学院硕士研究生,主要从事群智感知、云计算与物联网安全等方面的研究工作.
    • 基金资助:
    • 国家自然科学基金 (61403109); 高等学校博士学科点专项科研基金 (20112303120007); 黑龙江省自然科学基金 (LH2020F034)

Task Recommendation Method Based on Collaborative Ranking in Mobile Crowd Sensing

WANG Jian1, LIU Jia-xin1, ZHAO Guo-sheng2, ZHAO Zhong-nan1   

  1. 1.School of Computer Science and Technology, Harbin University of Science and Technology, Harbin, Heilongjiang 150080, China
    2.College of Computer Science and Information Engineering, Harbin Normal University, Harbin, Heilongjiang, 150025, China
  • Received:2020-02-28 Revised:2020-09-28 Online:2021-11-29 Published:2021-10-25
    • Supported by:
    • National Natural Science Foundation of China (61403109); Research Fund for the Doctoral Program of Higher Education of China (20112303120007); Natural Science Foundation of Heilongjiang Province, China (LH2020F034)

摘要:

针对移动群智感知中参与者积极性不高导致的数据质量低和激励成本高的问题,本文提出了一种基于混合用户模型与列表级排序学习算法相结合的协同排序任务推荐方法.根据参与者的历史行为对其进行分析,初步过滤掉一些劣质感知用户,同时利用参与者间的相似性构建混合用户模型.利用概率矩阵分解对参与者的意愿值进行预测,并根据排序学习得到一个排序模型.根据排序模型生成任务推荐列表,作为目标参与者的优选任务列表.基于真实数据集的仿真实验结果表明,本文提出的方法有效地提高了任务分配的准确率,与此同时减少了感知用户的移动距离.

关键词: 移动群智感知, 任务推荐, 协同排序, 混合用户模型, 参与者意愿

Abstract:

To solve the problem of low data quality and high incentive cost caused by the inactivity of participants in mobile crowd sensing, this paper proposes a task recommendation method. We could analyze the participants according to their historical behavior, and filter out low-quality sensing users. Meanwhile, the similarity among the participants was used to build a user-hybrid model. Then, the participants' willingness would be predicted by the probabilistic matrix factorization, and a ranking model was obtained. Finally, a task recommendation list was generated on the basis of ranking model as the preferred task list for the target participants. The simulation experiments based on the real dataset show that the proposed method in this paper can improve the accuracy of task assignment effectively and reduce the moving distance of sensing users simultaneously.

Key words: mobile crowd sensing, task recommendation, collaborative ranking, user-hybrid model, participants' willingness

中图分类号: