• 主题
高级搜索  >
历史搜索 清空历史
首页> 外文期刊>Big Earth Data >Enhancing VGI application semantics by accounting for spatial bias
【24h】

Enhancing VGI application semantics by accounting for spatial bias

机译：通过考虑空间偏差来增强VGI应用程序语义

获取原文
           
页面导航
  • 摘要
  • 著录项
  • 相似文献
  • 相关主题

摘要

Volunteered geographic information (VGI) is becoming an importantsource of geospatial big data that support many applications. Theapplication semantics of VGI, i.e. how well VGI reflects the real-worldgeographic phenomena of interest to the application, is essential forany VGI applications. VGI observations often are spatially biased (e.g.spatially clustered). Spatial bias poses challenges on VGI applicationsemantics because it may impede the quality of inferences madefrom VGI. Using species distribution modeling (SDM) as an exampleapplication, this article argues that spatial bias impedes VGI applica-tion semantics, asgauged by SDM model performance, and account-ing for bias enhances application semantics. VGI observations fromeBird were used in a case study for modeling the distribution of theAmerican Robin (Turdus migratorius) in U.S. T. migratorius observa-tions from the North American Breeding Bird Survey were used asindependent validation data for model performance evaluation.A grid-based strategy was adopted to filter eBird species observa-tions to reduce spatial bias. Evaluations show that spatial bias inspeciesobservations degradesSDM modelperformanceand filteringspecies observations improves model performance. This studydemonstrates that VGI application semantics can be enhanced byaccounting for the spatial bias in VGI observations.
机译：自愿性地理信息（VGI）成为支持许多应用程序的地理空间大数据的重要来源。 VGI的应用程序语义，即VGI反映应用程序感兴趣的真实地理现象的程度，对于任何VGI应用程序都是必不可少的。 VGI观测值通常在空间上有偏差（例如空间聚类）。空间偏差对VGI应用程序语义提出了挑战，因为它可能会阻碍从VGI进行推断的质量。本文以物种分布建模（SDM）为例，认为空间偏差会阻碍VGI应用语义，而SDM模型的性能会降低空间偏差，解决偏差会增强应用语义。案例研究中使用了来自eBird的VGI观测数据来对北美知更鸟观测中美国知更鸟（Turdus migratorius）的分布进行建模，并将其用作模型性能评估的独立验证数据，并采用了基于网格的策略过滤eBird物种的观测值以减少空间偏差。评估表明，空间偏差物种观测会降低SDM模型的性能，而过滤物种观测会提高模型的性能。这项研究表明，通过考虑VGI观测中的空间偏差，可以增强VGI应用程序的语义。

著录项

相似文献

  • 外文文献
  • 中文文献
  • 专利
获取原文

客服邮箱：kefu@zhangqiaokeyan.com

京公网安备：11010802029741号 ICP备案号：京ICP备15016152号-6 六维联合信息科技 (北京) 有限公司©版权所有

  • 客服微信

  • 服务号