A Review of Bayesian Networks for Spatial Data

IF 1.2 Q4 REMOTE SENSING

ACM Transactions on Spatial Algorithms and Systems Pub Date : 2022-03-30 DOI:10.1145/3516523

C. Krapu, R. Stewart, A. Rose

{"title":"A Review of Bayesian Networks for Spatial Data","authors":"C. Krapu, R. Stewart, A. Rose","doi":"10.1145/3516523","DOIUrl":null,"url":null,"abstract":"Bayesian networks are a popular class of multivariate probabilistic models as they allow for the translation of prior beliefs about conditional dependencies between variables to be easily encoded into their model structure. Due to their widespread usage, they are often applied to spatial data for inferring properties of the systems under study and also generating predictions for how these systems may behave in the future. We review published research on methodologies for representing spatial data with Bayesian networks and also summarize the application areas for which Bayesian networks are employed in the modeling of spatial data. We find that a wide variety of perspectives are taken, including a GIS-centric focus on efficiently generating geospatial predictions, a statistical focus on rigorously constructing graphical models controlling for spatial correlation, as well as a range of problem-specific heuristics for mitigating the effects of spatial correlation and dependency arising in spatial data analysis. Special attention is also paid to potential future directions for the integration of Bayesian networks with spatial processes.","PeriodicalId":43641,"journal":{"name":"ACM Transactions on Spatial Algorithms and Systems","volume":"9 1","pages":"1 - 21"},"PeriodicalIF":1.2000,"publicationDate":"2022-03-30","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"1","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"ACM Transactions on Spatial Algorithms and Systems","FirstCategoryId":"1085","ListUrlMain":"https://doi.org/10.1145/3516523","RegionNum":0,"RegionCategory":null,"ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"Q4","JCRName":"REMOTE SENSING","Score":null,"Total":0}

引用次数: 1

Abstract

Bayesian networks are a popular class of multivariate probabilistic models as they allow for the translation of prior beliefs about conditional dependencies between variables to be easily encoded into their model structure. Due to their widespread usage, they are often applied to spatial data for inferring properties of the systems under study and also generating predictions for how these systems may behave in the future. We review published research on methodologies for representing spatial data with Bayesian networks and also summarize the application areas for which Bayesian networks are employed in the modeling of spatial data. We find that a wide variety of perspectives are taken, including a GIS-centric focus on efficiently generating geospatial predictions, a statistical focus on rigorously constructing graphical models controlling for spatial correlation, as well as a range of problem-specific heuristics for mitigating the effects of spatial correlation and dependency arising in spatial data analysis. Special attention is also paid to potential future directions for the integration of Bayesian networks with spatial processes.

查看原文本刊更多论文

空间数据贝叶斯网络研究进展

贝叶斯网络是一类流行的多变量概率模型，因为它们允许将变量之间的条件依赖关系的先验信念转换为容易编码到它们的模型结构中。由于它们的广泛使用，它们经常被应用于空间数据，以推断所研究系统的性质，并对这些系统未来的行为产生预测。我们回顾了用贝叶斯网络表示空间数据的方法，并总结了贝叶斯网络在空间数据建模中的应用领域。我们发现采用了各种各样的视角，包括以gis为中心的高效生成地理空间预测，以统计学为中心的严格构建控制空间相关性的图形模型，以及一系列针对特定问题的启发式方法，以减轻空间数据分析中产生的空间相关性和依赖性的影响。特别关注贝叶斯网络与空间过程整合的潜在未来方向。

本文章由计算机程序翻译，如有差异，请以英文原文为准。

求助全文

约1分钟内获得全文求助全文

来源期刊

ACM Transactions on Spatial Algorithms and Systems REMOTE SENSING-

CiteScore

4.40

自引率

5.30%

发文量

期刊介绍： ACM Transactions on Spatial Algorithms and Systems (TSAS) is a scholarly journal that publishes the highest quality papers on all aspects of spatial algorithms and systems and closely related disciplines. It has a multi-disciplinary perspective in that it spans a large number of areas where spatial data is manipulated or visualized (regardless of how it is specified - i.e., geometrically or textually) such as geography, geographic information systems (GIS), geospatial and spatiotemporal databases, spatial and metric indexing, location-based services, web-based spatial applications, geographic information retrieval (GIR), spatial reasoning and mining, security and privacy, as well as the related visual computing areas of computer graphics, computer vision, geometric modeling, and visualization where the spatial, geospatial, and spatiotemporal data is central.