File Download

There are no files associated with this item.

  Links for fulltext
     (May Require Subscription)
Supplementary

Article: Geographically convolutional neural network weighted regression: a method for modeling spatially non-stationary relationships based on a global spatial proximity grid

TitleGeographically convolutional neural network weighted regression: a method for modeling spatially non-stationary relationships based on a global spatial proximity grid
Authors
KeywordsGeographically convolutional neural network weighted regression
geographically neural network weighted regression
geographically weighted regression
global spatial proximity grid
spatial non-stationarity
Issue Date2022
Citation
International Journal of Geographical Information Science, 2022, v. 36, n. 11, p. 2248-2269 How to Cite?
AbstractGeographically weighted regression (GWR) is a classical method of modeling spatially non-stationary relationships. The geographically neural network weighted regression (GNNWR) model solves the problem of the inaccurate construction of spatial weight kernels using a spatially weighted neural network. However, when the spatial distribution of observations is uneven, the spatial proximity expression in the input of GWR and GNNWR models does not fully represent the impact of the whole research space on the estimating point. Therefore, we established a global spatial proximity grid (GSPG) to express the spatial proximity of each estimating point and proposed a spatially weighted convolutional neural network (SWCNN) to extract the relationship between the GSPG and spatial weights. Finally, we proposed a geographically convolutional neural network weighted regression (GCNNWR) model combining SWCNN and ordinary linear regression (OLR) model to estimate spatial non-stationarity. We used two case studies of simulated data and real environment data to demonstrate the advancements of the GCNNWR model. The GCNNWR model achieved higher estimation accuracy and greater predictive power than the OLR, GWR, multi-scale GWR (MGWR), and GNNWR models. Moreover, the GCNNWR model maintained its better stability and accuracy in estimating spatially non-stationary relationships when the distribution of observations was uneven.
Persistent Identifierhttp://hdl.handle.net/10722/329882
ISSN
2023 Impact Factor: 4.3
2023 SCImago Journal Rankings: 1.436
ISI Accession Number ID

 

DC FieldValueLanguage
dc.contributor.authorDai, Zhen-
dc.contributor.authorWu, Sensen-
dc.contributor.authorWang, Yuanyuan-
dc.contributor.authorZhou, Hongye-
dc.contributor.authorZhang, Feng-
dc.contributor.authorHuang, Bo-
dc.contributor.authorDu, Zhenhong-
dc.date.accessioned2023-08-09T03:36:02Z-
dc.date.available2023-08-09T03:36:02Z-
dc.date.issued2022-
dc.identifier.citationInternational Journal of Geographical Information Science, 2022, v. 36, n. 11, p. 2248-2269-
dc.identifier.issn1365-8816-
dc.identifier.urihttp://hdl.handle.net/10722/329882-
dc.description.abstractGeographically weighted regression (GWR) is a classical method of modeling spatially non-stationary relationships. The geographically neural network weighted regression (GNNWR) model solves the problem of the inaccurate construction of spatial weight kernels using a spatially weighted neural network. However, when the spatial distribution of observations is uneven, the spatial proximity expression in the input of GWR and GNNWR models does not fully represent the impact of the whole research space on the estimating point. Therefore, we established a global spatial proximity grid (GSPG) to express the spatial proximity of each estimating point and proposed a spatially weighted convolutional neural network (SWCNN) to extract the relationship between the GSPG and spatial weights. Finally, we proposed a geographically convolutional neural network weighted regression (GCNNWR) model combining SWCNN and ordinary linear regression (OLR) model to estimate spatial non-stationarity. We used two case studies of simulated data and real environment data to demonstrate the advancements of the GCNNWR model. The GCNNWR model achieved higher estimation accuracy and greater predictive power than the OLR, GWR, multi-scale GWR (MGWR), and GNNWR models. Moreover, the GCNNWR model maintained its better stability and accuracy in estimating spatially non-stationary relationships when the distribution of observations was uneven.-
dc.languageeng-
dc.relation.ispartofInternational Journal of Geographical Information Science-
dc.subjectGeographically convolutional neural network weighted regression-
dc.subjectgeographically neural network weighted regression-
dc.subjectgeographically weighted regression-
dc.subjectglobal spatial proximity grid-
dc.subjectspatial non-stationarity-
dc.titleGeographically convolutional neural network weighted regression: a method for modeling spatially non-stationary relationships based on a global spatial proximity grid-
dc.typeArticle-
dc.description.naturelink_to_subscribed_fulltext-
dc.identifier.doi10.1080/13658816.2022.2100892-
dc.identifier.scopuseid_2-s2.0-85139128148-
dc.identifier.volume36-
dc.identifier.issue11-
dc.identifier.spage2248-
dc.identifier.epage2269-
dc.identifier.eissn1362-3087-
dc.identifier.isiWOS:000860607700001-

Export via OAI-PMH Interface in XML Formats


OR


Export to Other Non-XML Formats