山东大学学报 (工学版) ›› 2023, Vol. 53 ›› Issue (2): 93-101.doi: 10.6040/j.issn.1672-3961.0.2022.340
褚佳静,潘庆先*,潘亚楠,刘庆菊
CHU Jiajing, PAN Qingxian*, PAN Ya'nan, LIU Qingju
摘要： 针对目前众包平台会产生大量恶意工人以及较少考虑激励工人多次提供可信服务的问题,提出一种基于信誉模型的众包质量控制算法——信誉期望最大化(reputation expectation maximum, Rep-EM)算法。根据可信因子和惩罚因子建立信誉模型;基于工人信誉值和对任务的熟悉度提出一种工人选择机制;将工人匹配度作为权重赋予相应的工人并使用多数投票方法进行初始值选取,解决期望最大化(expectation maximum, EM)算法对初始值敏感和收敛困难的问题,避免算法陷入局部最优,提高评估结果的准确率;利用公开的众包数据集Adult2和Duck对Rep-EM算法和本研究提出的机制进行验证。试验结果表明,Rep-EM算法在评估准确率和运行时间方面有很大的提升,也从任务完成率和平均数据质量验证了本研究提出的工人选择机制的有效性。
中图分类号:
| [1] | HOWE J. The rise of crowdsourcing[J/OL]. WIRED.(2006-06-01)[2022-07-17]. https://www.wired.com/2006/06/crowds/. |
| [2] | OLESON D, SOROKIN A, LAUGHLIN G, et al. Programmatic gold: targeted and scalable quality assurance in crowdsourcing[C] //Proceedings of the 11th AAAI Conference on Human Computation. San Francisco, USA: AAAI Press, 2011: 43-48. |
| [3] | ZHAO D, MA H D, JI X N. Generalized lottery trees: budget-balanced incentive tree mechanisms for crowd-sourcing[J]. IEEE Transactions on Mobile Computing, 2021, 20(7): 2379-2394. |
| [4] | 熊金波, 马蓉, 牛犇, 等. 移动群智感知中基于用户联盟匹配的隐私保护激励机制[J]. 计算机研究与发展, 2018, 55(7): 1359-1370. XIONG Jinbo, MA Rong, NIU Ben, et al. Privacy pr-otection incentive mechanism based on user-union matching in mobile crowdsensing[J]. Journal of Computer Research and Development, 2018, 55(7): 1359-1370. |
| [5] | 冯景瑜, 杨锦雯, 张瑞通, 等. 抗位置隐私泄露的物联网频谱共享激励机制[J]. 计算机研究与发展, 2020, 57(10): 2209-2220. FENG Jingyu, YANG Jinwen, ZHANG Ruitong, et al. A spectrum sharing incentive scheme against location privacy leakage in IoT networks[J]. Journal of Computer Research and Development, 2020, 57(10): 2209-2220. |
| [6] | JIANG L Y, NIU X F, XU J, et al. Incentive mechanism design for truth discovery in crowdsourcing with copiers[J]. IEEE Transactions on Services Computing, 2021, 15(5): 2838-2853. |
| [7] | QIN H, WANG S L, MA P Z, et al. Quality control in crowdsourcing using sequential zero-determinant strategies[J]. IEEE Transactions on Knowledge and Data Engineering, 2020, 32(5): 998-1009. |
| [8] | AN J, WANG Z X, HE X, et al. PPQC: a blockchain-based privacy-preserving quality control mechanism in crowd-sensing applications[J]. IEEE/ACM Transactions on Net-working, 2022, 30(3): 1-16. |
| [9] | LIU X J, MA O B, CHEN W, et al. HDRS: a hybrid reputation system with dynamic update interval for detecting malicious vehicles in VANETs[J]. IEEE Transactions on Intelligent Transportation Systems, 2021, 23(8): 1-12. |
| [10] | 何云华, 李梦茹, 李红, 等. 群智感知应用中基于区块链的激励机制[J]. 计算机研究与发展, 2019, 56(3): 544-554. HE Yunhua, LI Mengru, LI Hong, et al. A blockchainbased incentive mechanism for crowdsensing applications[J]. Journal of Computer Research and Development, 2019, 56(3): 544-554. |
| [11] | 潘庆先, 江珊, 董红斌, 等. 基于Pareto分布的众包工人欺骗行为处理方法[J]. 计算机应用, 2019, 39(11): 3191-3197. PAN Qingxian, JIANG Shan, DONG Hongbin, et al. Processing method of crowdsourcing workers' cheating behavior based on Pareto distribution[J]. Computer Application, 2019, 39(11): 3191-3197. |
| [12] | CHOUIKHI S, KHOUKHI L, AVED S. An efficient reputation management model based on game theory for vehicular networks[C] //Proceedings of 2020 IEEE 45th Conference on Local Computer Networks. Sydney, Australia: IEEE, 2020: 413-416. |
| [13] | WANG Y J, GAO Y, LI Y S, et al. A worker-selection incentive mechanism for optimizing platform-centric mobile crowdsourcing systems[J]. Computer Networks, 2020, 171(22): 107144. |
| [14] | KANG J W, XIONG Z H, NIVATO D, et al. Incentive mechanism for reliable federated learning: a joint optimization approach to combining reputation and contract theory[J]. IEEE Internet of Things Journal, 2019, 6(6): 10700-10714. |
| [15] | 仲秋雁, 刘志娟. 考虑工作者信誉的众包质量EM评估方法[J]. 科技管理研究, 2018, 38(21): 70-76. ZHONG Qiuyan, LIU Zhijuan. EM evaluation method of crowdsourcing quality considering worker reputation[J]. Research on Science and Technology Management, 2018, 38(21): 70-76. |
| [16] | XU T Y, SUN H R, SU Z Y, et al. A quality assessment model for blockchain-based crowdsourcing system[C] //IEEE 24th International Conference on Computer Supported Cooperative Work in Design. Dalian, China: IEEE, 2021: 794-799. |
| [17] | YUEN M C, KING I, LEUNG K S. Task recommendation in crowdsourcing systems[C] //Proceedings of International Workshop on Crowdsourcing & Data Mining. Beijing, China: Association for Computing Machinery, 2012: 22-26. |
| [18] | 张宇航,周晓光,侯东阳. 综合多要素的地理空间数据众包任务推荐方法[J]. 测绘通报, 2022, 68(1): 84-88. ZHANG Yuhang, ZHOU Xiaoguang, HOU Dongyang. Recommendation method of geo spatial data crowdsourcing task based on multi elements[J]. Bulletin of Surveying and Mapping, 2022, 68(1): 84-88. |
| [19] | IPEIROTIS P, PROVOST F, WANG J. Quality management on amazon mechanical turk[C] //ACM SIGKDD Workshop on Human Computation(HCOMP'10). New York, USA: Association for Computing Machinery, 2010: 64-67. |
| [20] | WELINDER P, BRANSON S, BELONGIE S, et al. The multidimensional wisdom of crowds[C] //Proceedings of the 23rd International Conference on Neural Information Processing Systems: Volume 2. Vancouver, Canada: Curran Associates Inc, 2010: 2424-2432. |
| [21] | 芮兰兰, 张攀, 黄豪球, 等. 一种面向众包的基于信誉值的激励机制[J]. 电子与信息学报, 2016, 38(7): 1808-1815. RUI Lanlan, ZHANG Pan, HUANG Haoqiu, et al. Reputation based incentive mechanisms in crowdsourcing[J]. Journal of Electronics & Information Technology, 2016, 38(7): 1808-1815. |
| [22] | GAO Y L, LI X Y, LI J R, et al. A dynamic-trust-based recruiting framework for mobile crowd sensing[C] //Proceedings of 2017 IEEE International Conference on Communications(ICC). Paris, France: IEEE, 2017: 1-6. |
| [1] | 张琦,王莹洁. 时空众包下复合选择的长期激励算法[J]. 山东大学学报 (工学版), 2022, 52(2): 128-134. |
|
