移动众包环境下一种多阶段质量感知的在线激励机制  

Online Incentive Mechanism of Multi-stage Quality Perception in Mobile Crowdsourcing Environment

作　　者：高丽萍[1,2] 孙明达 高丽[3] 陈庆奎[1] GAO Li-ping;SUN Ming-da;GAO Li;CHEN Qing-kui(School of Optical-Electrical Computer Engineering,University of Shanghai for Science&Technology,Shanghai 200093,China;Shanghai Key Laboratory of Data Science,Fudan University,Shanghai 200093,China;Library Department of Shanghai University of Technology,Shanghai 200093,China)

机构地区：[1]上海理工大学光电信息与计算机工程学院,上海200093 [2]复旦大学上海数据科学重点实验室,上海200093 [3]上海理工大学图书馆,上海200093

出　　处：《小型微型计算机系统》2022年第5期1102-1108,共7页Journal of Chinese Computer Systems

基　　金：国家自然科学基金项目(61572325,60970012,61672354)资助;上海重点科技攻关项目(14511107902,16DZ1203603)资助;上海智能家居大规模物联共性技术工程中心项目(GCZX14014)资助。

摘　　要：随着移动人群感知成为收集数据的一种新范式,越来越多的移动众包应用应运而生.如何有效地激励工人参与任务并提供高质量的传感数据一直是该领域的关键问题,然而在线场景下的移动众包平台往往缺乏真实数据和未来信息,导致工人提交的数据质量参差不齐.所以本文考虑了在线场景下工人选择问题,提出一种多阶段质量感知的在线激励机制(SQOI).在预算和任务时限的约束下将整个活动周期分为多个阶段,每个阶段对上线的工人进行筛选采样、评估质量和更新参数,为下一阶段的选择策略提供依据.本文从理论上证明了该机制具有预算可行性、个体合理性、真实性和计算效率,同时,实验验证了该机制在提升数据质量上的有效性.As mobile crowd sensing becomes a new paradigm for data collection,more and more mobile crowdsourcing applications emerge.How to effectively motivate workers to participate in tasks and provide high quality sensor data has always been a key issue in this field.However,the mobile crowdsourcing platform under the online scene often lacks real data and future information,resulting in uneven data quality submitted by workers.Therefore,this paper considers the problem of worker selection in the online scenario and proposes a multi-stage quality perception online incentive mechanism(SQOI),which divides the entire activity cycle into several stages under the constraints of budget and task time limit.At each stage,the online workers are sampled,quality is evaluated and parameters are updated to provide a basis for the selection strategy of the next stage.This paper theoretically proves that the mechanism has budget feasibility,individual rationality,truthful and computational efficiency,and meanwhile,experiments verify the effectiveness of the mechanism in improving data quality.

关 键 词：移动众包 激励机制 质量控制 在线算法 

分 类 号：TP301[自动化与计算机技术—计算机系统结构]