Modeling of spatial stratified heterogeneity
| Autor: | Jiangang Guo, Jinfeng Wang, Chengdong Xu, Yongze Song |
|---|---|
| Jazyk: | angličtina |
| Rok vydání: | 2022 |
| Předmět: | |
| Zdroj: | GIScience & Remote Sensing, Vol 59, Iss 1, Pp 1660-1677 (2022) |
| Druh dokumentu: | article |
| ISSN: | 1548-1603 1943-7226 15481603 |
| DOI: | 10.1080/15481603.2022.2126375 |
| Popis: | Spatial stratified heterogeneity (SSH) refers to the geographical phenomena in which the geographical attributes within-strata are more similar than the between-strata, which is ubiquitous in the real world and offers information implying the causation of nature. Stratification, a primary approach to SSH, generates strata using a priori knowledge or thousands of supervised and unsupervised learning methods. Selecting reasonable stratification methods for spatial analysis in specific domains without prior knowledge is challenging because no method is optimal in a general sense. However, a systematic review of a large number of stratification methods is still lacking. In this article, we review the methods for stratification, categorize the existing typical stratification methods into four classes – univariate stratification, cluster-based stratification, multicriteria stratification, and supervised stratification – and construct their taxonomy. Finally, we present a summary of the software and tools used to compare and perform stratification methods. Given that different stratification methods reflect distinct human understandings of spatial distributions and associations, we suggest that further studies are needed to reveal the nature of geographical attributes by integrating SSH, advanced algorithms, and interdisciplinary methods. |
| Databáze: | Directory of Open Access Journals |
| Externí odkaz: |
|