基于消错决策理论的数据质量评估方法
|
摘要
针对移动群智感知中数据质量难以保障和评估的问题,从规避错误的角度出发,提出了一种基于消错决策理论的移动群智感知数据质量评估方法。通过引入消错决策理论来评估移动用户所提供的感知数据质量,对低质量及异常数据进行识别;考虑到不同感知任务具有不同的数据质量需求,引入权值因子实现数据质量的排序与优化评估。实验结果表明:所提出的方法不仅达到了对低质量和异常数据的准确识别的目标,还实现了对不同任务下的数据质量按需排序。在对低质量和异常数据识别方面,相对于拉依达准则,具有更高的识别准确性;在对数据质量排序上,相对于理想点法、TOPSIS法,具有准确性高和简单高效的优点。
In mobile crowd-sensing scenarios,it is difficult to guarantee and evaluate the quality of data collected from a number of mobile users. To solve this problem,we proposed a novel technique for evaluating the quality of crowd-sensing data based on the error-eliminating decision-making theory with regard to error avoidance as the objective. First,we evaluated the quality of data collected from the mobile users by introducing the error-eliminating decision-making theory,wherein low-quality and abnormal data can be identified. Then,in consideration with various requirements regarding data quality for different sensing tasks,we introduced a weight factor for sequencing and evaluating data quality. Our experimental results show that the proposed method can accurately identify low-quality and abnormal data as well as achieve the on-demand sequencing of data quality with respect to different task requirements. In terms of identifying low-quality and abnormal data,our proposed method demonstrates higher recognition accuracy than the Laidda criterion. Additionally,as compared with the ideal-point and TOPSIS methods,our proposed method exhibits high accuracy and simplicity,proving to be advantageous in terms of data quality ranking.
In mobile crowd-sensing scenarios,it is difficult to guarantee and evaluate the quality of data collected from a number of mobile users. To solve this problem,we proposed a novel technique for evaluating the quality of crowd-sensing data based on the error-eliminating decision-making theory with regard to error avoidance as the objective. First,we evaluated the quality of data collected from the mobile users by introducing the error-eliminating decision-making theory,wherein low-quality and abnormal data can be identified. Then,in consideration with various requirements regarding data quality for different sensing tasks,we introduced a weight factor for sequencing and evaluating data quality. Our experimental results show that the proposed method can accurately identify low-quality and abnormal data as well as achieve the on-demand sequencing of data quality with respect to different task requirements. In terms of identifying low-quality and abnormal data,our proposed method demonstrates higher recognition accuracy than the Laidda criterion. Additionally,as compared with the ideal-point and TOPSIS methods,our proposed method exhibits high accuracy and simplicity,proving to be advantageous in terms of data quality ranking.
引文
[1]ZHANG Xinglin,YANG Zheng,SUN Wei,et al. Incentives for mobile crowd sensing:a survey[J]. IEEE communications surveys&tutorials,2016,18(1):54-67.
[2]ALVEAR O,CALAFATE C,CANO J C,et al. Crowdsensing in smart cities:overview,platforms,and environment sensing issues[J]. Sensors,2018,18(2):460.
[3]WANG Xiong,ZHANG Jinbei,TIAN Xiaohua,et al.Crowdsensing-based consensus incident report for road traffic acquisition[J]. IEEE transactions on intelligent transportation systems,2018,19(8):2536-2547.
[4]GUO Bin,WANG Zhu,YU Zhiwen,et al. Mobile crowd sensing and computing:The review of an emerging humanpowered sensing paradigm[J]. ACM computing surveys,2015,48(1):7.
[5]JIANG Changkun,GAO Lin,DUAN Lingjie,et al. Exploiting data reuse in mobile crowdsensing[C]//2016 IEEE Global Communications Conference. Washington, DC,USA,2017:1-6.
[6]DU Pengfei,YANG Qinghai,HE Qingsu,et al. Energy-aware quality of information maximisation for wireless sensor networks[J]. IET communications,2016,10(17):2281-2289.
[7]吴垚,曾菊儒,彭辉,等.群智感知激励机制研究综述[J].软件学报,2016,27(8):2025-2047.WU Yao,ZENG Juru,PENG Hui,et al. Survey on incentive mechanisms for crowd sensing[J]. Journal of software,2016,27(8):2025-2047.
[8]南文倩,郭斌,陈荟慧,等.基于跨空间多元交互的群智感知动态激励模型[J].计算机学报,2015,38(12):2412-2425.NAN Wenqian,GUO Bin,CHEN Huihui,et al. A crossspace,multi-interaction-based dynamic incentive mechanism for mobile crowd sensing[J]. Chinese journal of computers,2015,38(12):2412-2425.
[9]PENG Dan,WU Fan,CHEN Guihai. Data quality guided incentive mechanism design for crowdsensing[J]. IEEE transactions on mobile computing,2018,17(2):307-319.
[10]ZHOU Tongqing,CAI Zhiping,CHEN Yueyue,et al. Improving data credibility for mobile crowdsensing with clustering and logical reasoning[C]//Second International Conference on Cloud Computing and Security. Nanjing,2016:138-150.
[11]LIU Shengzhong,ZHENG Zhenzhe,WU Fan,et al. Context-aware data quality estimation in mobile crowdsensing[C]//IEEE INFOCOM 2017-IEEE Conference on Computer Communications. Atlanta,GA,USA,2017:1-9.
[12]SUROWIECKI J. The wisdom of crowds:Why the many are smarter than the few and how collective wisdom shapes business,economics,societies and nations[J]. Personnel psychology,2006,59(4):982-985.
[13]MALONE T W,LAUBACHER R,DELLAROCAS C.Harnessing crowds:mapping the genome of collective intelligence. MIT Sloan Research Paper No. 4732-09[R].Cambridge, Massachusetts:Massachusetts Institute of Technology,2009.
[14]DOAN A, RAMAKRISHNAN R, HALEVY A Y.Crowdsourcing systems on the world-wide web[J]. Communications of the ACM,2011,54(4):86-96.
[15]LANE N D,MILUZZO E,LU Hong,et al. A survey of mobile phone sensing[J]. IEEE communications magazine,2010,48(9):140-150.
[16]HUANG Haoran,CAI Ken. A method of fuzzy multiple attribute decision making based on the error-eliminating theory[J]. Journal of intelligent&fuzzy systems,2016,31(4):2119-2127.
[17]MARJANOVIC'M,SKORIN-KAPOV L,PRIPUIC'K,et al. Energy-aware and quality-driven sensor management for green mobile crowd sensing[J]. Journal of network and computer applications,2016,59:95-108.
[18]丁小欧,王宏志,张笑影,等.数据质量多种性质的关联关系研究[J].软件学报,2016,27(7):1626-1644.DING Xiaoou,WANG Hongzhi,ZHANG Xiaoying,et al.Association relationships study of multi-dimensional data quality[J]. Journal of software,2016,27(7):1626-1644.
[2]ALVEAR O,CALAFATE C,CANO J C,et al. Crowdsensing in smart cities:overview,platforms,and environment sensing issues[J]. Sensors,2018,18(2):460.
[3]WANG Xiong,ZHANG Jinbei,TIAN Xiaohua,et al.Crowdsensing-based consensus incident report for road traffic acquisition[J]. IEEE transactions on intelligent transportation systems,2018,19(8):2536-2547.
[4]GUO Bin,WANG Zhu,YU Zhiwen,et al. Mobile crowd sensing and computing:The review of an emerging humanpowered sensing paradigm[J]. ACM computing surveys,2015,48(1):7.
[5]JIANG Changkun,GAO Lin,DUAN Lingjie,et al. Exploiting data reuse in mobile crowdsensing[C]//2016 IEEE Global Communications Conference. Washington, DC,USA,2017:1-6.
[6]DU Pengfei,YANG Qinghai,HE Qingsu,et al. Energy-aware quality of information maximisation for wireless sensor networks[J]. IET communications,2016,10(17):2281-2289.
[7]吴垚,曾菊儒,彭辉,等.群智感知激励机制研究综述[J].软件学报,2016,27(8):2025-2047.WU Yao,ZENG Juru,PENG Hui,et al. Survey on incentive mechanisms for crowd sensing[J]. Journal of software,2016,27(8):2025-2047.
[8]南文倩,郭斌,陈荟慧,等.基于跨空间多元交互的群智感知动态激励模型[J].计算机学报,2015,38(12):2412-2425.NAN Wenqian,GUO Bin,CHEN Huihui,et al. A crossspace,multi-interaction-based dynamic incentive mechanism for mobile crowd sensing[J]. Chinese journal of computers,2015,38(12):2412-2425.
[9]PENG Dan,WU Fan,CHEN Guihai. Data quality guided incentive mechanism design for crowdsensing[J]. IEEE transactions on mobile computing,2018,17(2):307-319.
[10]ZHOU Tongqing,CAI Zhiping,CHEN Yueyue,et al. Improving data credibility for mobile crowdsensing with clustering and logical reasoning[C]//Second International Conference on Cloud Computing and Security. Nanjing,2016:138-150.
[11]LIU Shengzhong,ZHENG Zhenzhe,WU Fan,et al. Context-aware data quality estimation in mobile crowdsensing[C]//IEEE INFOCOM 2017-IEEE Conference on Computer Communications. Atlanta,GA,USA,2017:1-9.
[12]SUROWIECKI J. The wisdom of crowds:Why the many are smarter than the few and how collective wisdom shapes business,economics,societies and nations[J]. Personnel psychology,2006,59(4):982-985.
[13]MALONE T W,LAUBACHER R,DELLAROCAS C.Harnessing crowds:mapping the genome of collective intelligence. MIT Sloan Research Paper No. 4732-09[R].Cambridge, Massachusetts:Massachusetts Institute of Technology,2009.
[14]DOAN A, RAMAKRISHNAN R, HALEVY A Y.Crowdsourcing systems on the world-wide web[J]. Communications of the ACM,2011,54(4):86-96.
[15]LANE N D,MILUZZO E,LU Hong,et al. A survey of mobile phone sensing[J]. IEEE communications magazine,2010,48(9):140-150.
[16]HUANG Haoran,CAI Ken. A method of fuzzy multiple attribute decision making based on the error-eliminating theory[J]. Journal of intelligent&fuzzy systems,2016,31(4):2119-2127.
[17]MARJANOVIC'M,SKORIN-KAPOV L,PRIPUIC'K,et al. Energy-aware and quality-driven sensor management for green mobile crowd sensing[J]. Journal of network and computer applications,2016,59:95-108.
[18]丁小欧,王宏志,张笑影,等.数据质量多种性质的关联关系研究[J].软件学报,2016,27(7):1626-1644.DING Xiaoou,WANG Hongzhi,ZHANG Xiaoying,et al.Association relationships study of multi-dimensional data quality[J]. Journal of software,2016,27(7):1626-1644.