空间大数据中的不确定性分析与数据质量控制
发布时间：2021-05-08 18:00
自“大数据”的概念在2005年被提出以来,“大数据”在全球范围内的学术界和商业界掀起一股革新的热潮。空间大数据（SBD）特指具有空间信息的大数据。智能技术和传感器技术的快速发展使得空间大数据成为了目前最具价值的空间产品之一。SBD可以被分为对地观测大数据与人类活动大数据两类。目前,运用先进的技术手段从SBD中提炼知识已经成为学术界以及相关产业中的重要范式;SBD也在地球科学,城市学,环境科学等多个科学领域持续推动着创新。不确定性被普遍认为是存在于空间数据研究中最重要的元素之一。不确定性广泛存在于空间数据生产和分析的全过程当中。对不确定性的不当处理可能会得到错误的结论并且造成巨大的损失。在过去的几十年间,针对空间数据以及空间分析的不确定性理论以及方法研究已经取得了显著进展,并在近年来被进一步扩展至空间数据与空间分析的可靠性研究中以满足应用层级对于可靠性的广泛需求。由于SBD庞大的数据量,涉及众多的数据类型以及复杂的不确定性问题,传统的不确定性分析方法,尤其是涉及大量人力,辅助信息以及个人判断的方法,在应对SBD不确定性问题时变得低效甚至无效。针对这一问题,本文旨在利用数据挖掘技术,提出一...
【文章来源】：武汉大学湖北省 211工程院校 985工程院校 教育部直属院校
【文章页数】：140 页
【学位级别】：博士
【文章目录】：
学位论文创新点
摘要
ABSTRACT
1 Introduction
1.1 Background and motivation
1.2 State of the art on SBD uncertainty analysis
1.3 Scope and objectives of this study
1.4 Structure of dissertation
2 Reference-free method for locating potential errors based on spatial relationship complexity in multilayer vector data
2.1 Overview of spatial relationship complexity
2.2 Complexity contribution from a single layer
2.3 Combined complexity from multiple layers
2.4 Experiments and analysis
2.5 Summary
3 Reference-free method for detecting classification errors in LULC big data
3.1 Overview of reference-free methods for LULC data evaluation
3.2 Detecting outlying features in land cover dataset
3.3 Experiments and analysis
3.4 Summary
4 Reference-free method for investigating scale uncertainty in LULC big data
4.1 Overview of scale uncertainty in LULC
4.2 Modelling omission errors
4.3 Modelling commission errors
4.4 Experiments and analysis
4.5 Summary
5 Adaptive uncertainty models for trajectory big data
5.1 Trajectory uncertainty models
5.2 Construction of AEE model
5.3 Construction of BAEE model
5.4 Experiments and analysis
5.5 Summary
6 Exploring the uncertainty in time-series big data and its application in human behaviour prediction
6.1 Overview of the predictability of time-series data
6.2 Modelling the predictability of human mobility as a time series
6.3 Experiments and analysis
6.4 Summary
7 Conclusions and recommendations
7.1 Summary of this thesis
7.2 Limitations
7.3 Recommendations
Appendix A Derivation process of the approximated ellipse for BAEE model
References
攻读博士期间的论文发表和科研工作
致谢
【参考文献】：
期刊论文
[1]地理国情监测可靠性分析[J]. 史文中,陈鹏飞,张效康. 测绘学报. 2017(10)
[2]可靠性空间分析初探[J]. 史文中,陈江平,詹庆明,舒红. 武汉大学学报(信息科学版). 2012(08)
[3]顾及地图要素级别的几何信息量量测方法[J]. 王少一,王昭,杜清运. 测绘科学. 2007(04)
本文编号：3175753
【文章来源】：武汉大学湖北省 211工程院校 985工程院校 教育部直属院校
【文章页数】：140 页
【学位级别】：博士
【文章目录】：
学位论文创新点
摘要
ABSTRACT
1 Introduction
1.1 Background and motivation
1.2 State of the art on SBD uncertainty analysis
1.3 Scope and objectives of this study
1.4 Structure of dissertation
2 Reference-free method for locating potential errors based on spatial relationship complexity in multilayer vector data
2.1 Overview of spatial relationship complexity
2.2 Complexity contribution from a single layer
2.3 Combined complexity from multiple layers
2.4 Experiments and analysis
2.5 Summary
3 Reference-free method for detecting classification errors in LULC big data
3.1 Overview of reference-free methods for LULC data evaluation
3.2 Detecting outlying features in land cover dataset
3.3 Experiments and analysis
3.4 Summary
4 Reference-free method for investigating scale uncertainty in LULC big data
4.1 Overview of scale uncertainty in LULC
4.2 Modelling omission errors
4.3 Modelling commission errors
4.4 Experiments and analysis
4.5 Summary
5 Adaptive uncertainty models for trajectory big data
5.1 Trajectory uncertainty models
5.2 Construction of AEE model
5.3 Construction of BAEE model
5.4 Experiments and analysis
5.5 Summary
6 Exploring the uncertainty in time-series big data and its application in human behaviour prediction
6.1 Overview of the predictability of time-series data
6.2 Modelling the predictability of human mobility as a time series
6.3 Experiments and analysis
6.4 Summary
7 Conclusions and recommendations
7.1 Summary of this thesis
7.2 Limitations
7.3 Recommendations
Appendix A Derivation process of the approximated ellipse for BAEE model
References
攻读博士期间的论文发表和科研工作
致谢
【参考文献】：
期刊论文
[1]地理国情监测可靠性分析[J]. 史文中,陈鹏飞,张效康. 测绘学报. 2017(10)
[2]可靠性空间分析初探[J]. 史文中,陈江平,詹庆明,舒红. 武汉大学学报(信息科学版). 2012(08)
[3]顾及地图要素级别的几何信息量量测方法[J]. 王少一,王昭,杜清运. 测绘科学. 2007(04)
本文编号：3175753
本文链接：https://www.wllwen.com/kejilunwen/ruanjiangongchenglunwen/3175753.html
